Prediction: Breaking $500 within 24 hours

[K128kevin]

Some of you have seen my website which tries to predict bitcoin prices with software using neural networks:

www.btcpredictions.com

It successfully predicted the drop in price that happened yesterday morning (EST) and it was very accurate throughout the day. It also predicted this rise that we are seeing now, and it's saying that prices will go into the upper $400s, possibly hitting $500! Here is an image of the predictions as of 11am EST:



The software is not always this accurate, but it has been performing exceptionally well recently! I think that we will see prices over $500 within the next 24 hours.

[1714998347]

1714998347

[1714998347]

1714998347

[1714998347]

1714998347

[1714998347]

1714998347

[spazzdla]

Hum... currently 449... hum.. I wouldn't bet against you.  Although I'd bet dollars to donuts china releases 2-4 bad news stories today in another attempt to drive the price down.

[zolace]

Hope your prediction is good!We need to see some rise on bitcoin value,some positive news ,not only negative ones.

[bryant.coleman]

It might break $500 for a few minutes, but will end up at 450-460 levels.

[binderclip]

Nice work - do you apply this methodology to other markets, like stocks and bonds?

[K128kevin]

Nice work - do you apply this methodology to other markets, like stocks and bonds?

I tried to do a dow jones prediction using a neural network, but the problem was that I was having trouble finding the most up to date data. I might revisit stock price predictions in the future though.

[K128kevin]

Looks like it changed its mind about hitting $500 =/ might just get to upper $400s. We'll find out soon.

[iraszl]

Cool project. Have you mapped previously predicted data to real data? Do you for example have a prediction from 20 days ago to check how it matched reality?

Posted about your project here: http://bitcoinowl.com/bitcoin-price-prediction

[PolarPoint]

So your neutral network predicts future trends based on past and current market behaviour?
How far into the past does it consider relevent?

[K128kevin]

Cool project. Have you mapped previously predicted data to real data? Do you for example have a prediction from 20 days ago to check how it matched reality?

Posted about your project here: http://bitcoinowl.com/bitcoin-price-prediction

Thanks for sharing it! I am planning on starting to collect/generate/display data for charts that will show predictions against actual prices. I am going to start that either tonight or tomorrow some time. A lot of people have been asking for something like that, and I think it would be interesting to see.

So your neutral network predicts future trends based on past and current market behaviour?
How far into the past does it consider relevent?

Yes, it trains on the historic bitstamp transaction data, which consists of all transactions for the past (about) 3 years. In order to determine what future prices will be, it looks at a sample of data leading up to the current price. So for example, the 24 hour prediction neural network looks at 60 average hourly prices in a row and tries to predict the next 24.

I experimented a lot with different distances into the past that it would look to make predictions. I tried having it just look at one day's data to predict the next, and I tried having it look at like 30 days worth of data to predict 1 day in the future, and I found that around 60 hours tended to work the best for the 24 hour one. The 5 day one looks at about 20 days, and the 20 day one looks at the past 150 days.

[kashish948]

I think it will break 5000

[bryant.coleman]

Still at 430-440 levels. Unfortunately, your prediction has failed. 

[K128kevin]

Still at 430-440 levels. Unfortunately, your prediction has failed. 

It's been only 10 hours, not 24. Plus it predicted this dip along the way, so actually the prediction so far has been very accurate and extremely successful

It has lowered the predicted peak price, but we'll see what happens.

[binderclip]

Nice work - do you apply this methodology to other markets, like stocks and bonds?

I tried to do a dow jones prediction using a neural network, but the problem was that I was having trouble finding the most up to date data. I might revisit stock price predictions in the future though.

You'd probably have to pay to get the up to date data.

[HCLivess]

I have been analysing the market since January and this overlaps with what I expect.
Perfect tool, thank you!

[softron]

So whats the forecast for the month.

[K128kevin]

I have been analysing the market since January and this overlaps with what I expect.
Perfect tool, thank you!

Really? That's awesome! I'm going to set it up today to start saving predictions and match them up against historic data on a chart. What method(s) have you been using to analyze the market?

[Cheesus]

So whats the forecast for the month.

Cloudy with possible bitcoinfall :>

[5thStreetResearch]

Ummm can we set up an escrow and wager on this?  Will give reasonable odds.

[redwhite037]

Well, so much for that prediction, looks like we're sideways trading between 430 and 450, but probably going to re-test 400 any day now. I'm looking to buy at 410 and 360.

[twiifm]

Any trader knows its impossible to predict the future.

Look up Random Walk Theory

[K128kevin]

Any trader knows its impossible to predict the future.

Look up Random Walk Theory

It's not impossible and people do it all the time. People make ridiculous amounts of money by predicting stock price movement seconds ahead of time and making super short term trades, seconds apart. This is just trying to do it on a longer term scale. And it seems to be working pretty well. Definitely better than any human could do.

[paulsonnumismatics]

It's not impossible and people do it all the time. People make ridiculous amounts of money by predicting stock price movement seconds ahead of time and making super short term trades, seconds apart. This is just trying to do it on a longer term scale. And it seems to be working pretty well. Definitely better than any human could do.

It is impossible. And has been demonstrated repeteadly and using different approaches and models. In a 30 year span a coin toss model has the same ROI than the 99.99% Hedge funds.
Then there is Warren Bufett. 

Don't mistake the skill and discipline that takes to anticipate tendencies with the magic power of making predictions.

Read this last sentence again.

Bingo ;-)

[K128kevin]

It is impossible. And has been demonstrated repeteadly and using different approaches and models. In a 30 year span a coin toss model has the same ROI than the 99.99% Hedge funds.
Then there is Warren Bufett. 

Don't mistake the skill and discipline that takes to anticipate tendencies with the magic power of making predictions.

Read this last sentence again.

Bingo ;-)

For you to argue that it is impossible to anticipate price changes using software is like arguing that it is impossible to go to the moon. You honestly could make a more convincing argument that it is impossible to go to the moon. You see here right in front of you, on my website, that I - as an undergrad studying computer science, far from a financial expert - was able to do it myself. This means that obviously people with much more experience than me can do it. And they do it on a regular basis. Use Google... you'll see that these types of strategies are commonly employed and proven to be successful.

Also honestly I find it hard to believe that anybody can believe that THIS is impossible. After all that humanity has done, all of our complex and great technological achievements, you think that predicting stock prices with reasonable accuracy is impossible? It's very possible and I have done it. So have many others.

[twiifm]

Any trader knows its impossible to predict the future.

Look up Random Walk Theory

It's not impossible and people do it all the time. People make ridiculous amounts of money by predicting stock price movement seconds ahead of time and making super short term trades, seconds apart. This is just trying to do it on a longer term scale. And it seems to be working pretty well. Definitely better than any human could do.

Who are these people?  You mean HFT?  They are front running trades to capture the spread.  It's similar to what a Market Maker does.

My friend works at an HFT shop in NJ.  All their positions are delta hedged.  They don't make money on predicting the future.  They make money on rebates & capturing the spread

If you claim you can predict then why is BTC not $500?

[K128kevin]

If you claim you can predict then why is BTC not $500?

This is like if a sharpshooter claims that they are pretty good and you ask them "If you are pretty good, then why can't you hit a mosquito's left eyeball from 2,000 meters away with a blindfold on?"

It's not perfect, as I've said many, many times. It is sometimes wrong, which is why it readjusts its predictions every hour. If you read the older posts you would have seen that it readjusted its prediction one hour after I posted this, saying that it would not reach $500. The neural network successfully predicted a pretty significant rise in price.

It doesn't have to be 100% perfect. I was arguing against the ridiculous claim that it is literally impossible to predict bitcoin prices. I'll give you that it's impossible to predict bitcoin prices 100% correctly 100% of the time, but it's very possible to predict them correctly within a reasonable margin of error the majority of the time. This is better than most (probably any, actually) humans could do.

[twiifm]

If you claim you can predict then why is BTC not $500?

This is like if a sharpshooter claims that they are pretty good and you ask them "If you are pretty good, then why can't you hit a mosquito's left eyeball from 2,000 meters away with a blindfold on?"

It's not perfect, as I've said many, many times. It is sometimes wrong, which is why it readjusts its predictions every hour. If you read the older posts you would have seen that it readjusted its prediction one hour after I posted this, saying that it would not reach $500. The neural network successfully predicted a pretty significant rise in price.

It doesn't have to be 100% perfect. I was arguing against the ridiculous claim that it is literally impossible to predict bitcoin prices. I'll give you that it's impossible to predict bitcoin prices 100% correctly 100% of the time, but it's very possible to predict them correctly within a reasonable margin of error the majority of the time. This is better than most (probably any, actually) humans could do.

I'm saying its impossible to mathematically model markets because they are based on human interactions which are totally unpredictable.

The one thing I find "predictable" is that implied volatility reverts to mean.  Not really a prediction.  Just that statistically speaking, volatility is mean reverting (not prices)

How are you predicting price?  What is your model based on?

[K128kevin]

I'm saying its impossible to mathematically model markets because they are based on human interactions which are totally unpredictable.

The one thing I find "predictable" is that implied volatility reverts to mean.  Not really a prediction.  Just that statistically speaking, volatility is mean reverting (not prices)

How are you predicting price?  What is your model based on?

The thing that you have to understand is that just because we, as humans, can't see predictable patterns in a set of data doesn't mean that they aren't there. Neural networks are REALLY good at finding patterns that humans cannot see. The first time I ever implemented a neural network was to recognize hand-written digits. It was able to correctly classify some hand written digits that were written so poorly that I, a human, could not tell for sure if they were one digit or another.

Human behavior is not totally unpredictable. Especially given large numbers of humans. This is besides the point though. The actual cause of price fluctuations is completely irrelevant actually. My computational model has no idea what is causing prices to change because it doesn't matter at all. All that matters is that there are repetitive patterns in the data (which there are). The neural network learns these patterns, and is able to use them to accurately predict bitcoin prices.

You can say you don't believe that bitcoin prices can be mathematically modeled, but it is still true - it's exactly what I'm doing (relatively successfully) here.

[Equate]

Don't know if you right but $500 with in 24 hrs is doesn't looks promising.

[twiifm]

I'm saying its impossible to mathematically model markets because they are based on human interactions which are totally unpredictable.

The one thing I find "predictable" is that implied volatility reverts to mean.  Not really a prediction.  Just that statistically speaking, volatility is mean reverting (not prices)

How are you predicting price?  What is your model based on?

The thing that you have to understand is that just because we, as humans, can't see predictable patterns in a set of data doesn't mean that they aren't there. Neural networks are REALLY good at finding patterns that humans cannot see. The first time I ever implemented a neural network was to recognize hand-written digits. It was able to correctly classify some hand written digits that were written so poorly that I, a human, could not tell for sure if they were one digit or another.

Human behavior is not totally unpredictable. Especially given large numbers of humans. This is besides the point though. The actual cause of price fluctuations is completely irrelevant actually. My computational model has no idea what is causing prices to change because it doesn't matter at all. All that matters is that there are repetitive patterns in the data (which there are). The neural network learns these patterns, and is able to use them to accurately predict bitcoin prices.

You can say you don't believe that bitcoin prices can be mathematically modeled, but it is still true - it's exactly what I'm doing (relatively successfully) here.

Human behavior predictable at the extreme.   I agree on that.   Large spikes always follow by selloff.  But without extremes its random.

I trade every day and I watch tick for tick on some stocks I trade.   Yes there are patterns but the pattern presents itself after each candle is drawn.   Theres always a lag that makes it useless as a predictor.

I love trading breakout patterns.   Breakouts occur after consolidation.   But I don't know direction of  breakout or to what price.

Its not that I don't believe you.   I've never met anyone who claim they can predict next days price.

I hope you become filthy rich and teach me your system

[K128kevin]

Human behavior predictable at the extreme.   I agree on that.   Large spikes always follow by selloff.  But without extremes its random.

I trade every day and I watch tick for tick on some stocks I trade.   Yes there are patterns but the pattern presents itself after each candle is drawn.   Theres always a lag that makes it useless as a predictor.

I love trading breakout patterns.   Breakouts occur after consolidation.   But I don't know direction of  breakout or to what price.

Its not that I don't believe you.   I've never met anyone who claim they can predict next days price.

I hope you become filthy rich and teach me your system

You should search Google for studies using neural networks for stock price prediction. You will find plenty of examples of it being done successful.

I will reiterate my main point that I expressed before though: Neural networks can find patterns that humans can't. Just because we can't see patterns or understand them doesn't mean that they aren't there. Neural networks can find complex and extensive patterns that no human could possibly ever understand. I use 60 inputs to predict the 24 prices for the 24 hour prediction. The hidden layer has 200 inputs. This basically is like representing a function with 12,000 variables. Obviously this is extremely complicated, and obviously there is no way a human could understand a function like this.

No matter how often you trade and how much time you spend trying to understand patterns in price fluctuations, you will never come even remotely close to being able to understand these patterns as thoroughly as a properly implemented neural network. It's like trying to beat a calculator at math - you can't do it.

[freedomno1]

Well I will agree that this chart is plausible so no real reason to disagree with it yet
All we can do is say we will see

[Easy2Mine]

@Kevin
Thanks for adding something new to BTC.
You did a terrible good job.
You are ahead of time for most people to understand how things works.
Most people also never heard of high frequency trading, which has already proven to work succesfully.

[SportsBet]

been following your predictions, they have been consistently off

[K128kevin]

@Kevin
Thanks for adding something new to BTC.
You did a terrible good job.
You are ahead of time for most people to understand how things works.
Most people also never heard of high frequency trading, which has already proven to work succesfully.

Thanks, I'm glad you can actually appreciate it

been following your predictions, they have been consistently off

hmmm, you obviously haven't been following them then. The only time they were consistently off during the past month was for about a 24-36 ish hours immediately after the China ban. They've been wrong before at other times but definitely not consistently.

[twiifm]

Human behavior predictable at the extreme.   I agree on that.   Large spikes always follow by selloff.  But without extremes its random.

I trade every day and I watch tick for tick on some stocks I trade.   Yes there are patterns but the pattern presents itself after each candle is drawn.   Theres always a lag that makes it useless as a predictor.

I love trading breakout patterns.   Breakouts occur after consolidation.   But I don't know direction of  breakout or to what price.

Its not that I don't believe you.   I've never met anyone who claim they can predict next days price.

I hope you become filthy rich and teach me your system

You should search Google for studies using neural networks for stock price prediction. You will find plenty of examples of it being done successful.

I will reiterate my main point that I expressed before though: Neural networks can find patterns that humans can't. Just because we can't see patterns or understand them doesn't mean that they aren't there. Neural networks can find complex and extensive patterns that no human could possibly ever understand. I use 60 inputs to predict the 24 prices for the 24 hour prediction. The hidden layer has 200 inputs. This basically is like representing a function with 12,000 variables. Obviously this is extremely complicated, and obviously there is no way a human could understand a function like this.

No matter how often you trade and how much time you spend trying to understand patterns in price fluctuations, you will never come even remotely close to being able to understand these patterns as thoroughly as a properly implemented neural network. It's like trying to beat a calculator at math - you can't do it.

Describe what you are doing and ill tell you if human is better or not.   HFT does not predict.   HFT front run orders and capture the spread.   It can do this better humans cause it can do it faster.   

But humans program it to do this.

I dont know if a machine can find patterns better than humans.    Give me a real world example

Let's do a test.   We guess the open close price each day of S&P for next 10 days.  If you're right you get a point.   Whoever scores the most point wins.   Just an experiment to see if your algo can guess the price better than me.   

At market close you have to guess the open price.   At open you guess the closing price.  Your allowed to be off by +/- 50c

[K128kevin]

Describe what you are doing and ill tell you if human is better or not.   HFT does not predict.   HFT front run orders and capture the spread.   It can do this better humans cause it can do it faster.   

But humans program it to do this.

I dont know if a machine can find patterns better than humans.    Give me a real world example

Let's do a test.   We guess the open close price each day of S&P for next 10 days.  If you're right you get a point.   Whoever scores the most point wins.   Just an experiment to see if your algo can guess the price better than me.   

At market close you have to guess the open price.   At open you guess the closing price.  Your allowed to be off by +/- 50c

I'm sorry but I'm not going to bother trying to get my software to predict S&P right now because I have a ton of other work to do. If you want to test accuracy, you should just make your own bitcoin price predictions and test their accuracy compared to my charts on this website.

If you want to know how it works look up artificial neural networks on Google. This isn't some simple process that I can explain to you in one post. It's a very complex computational model that people have been developing and researching since the 1940s. You could teach multiple college courses just on neural networks alone. To be honest, it seems like you are refuting 70 years worth of research based on your gut feeling that humans would be better at this - and if we're going to be realistic, computers are better than humans at nearly all tasks involving any type of computation. They're surpassing us quickly in other realms as well. They have been able to beat us in chess for a long time now, they can beat us in almost any other game, and they can even drive cars better than us now. Is it so hard to believe that they can predict stock prices better than us?

[MoreBloodWine]

and skynet is born !

C'mon people, if it's right it's right, if it's wrong it's wrong... ether way just have fun.

[jamesc760]

Kevin,

Thank you for the wonderful tool you have given us. I just hope you will continue to do so.

At this point in time, I think it's almost a moot point how accurate kevin's neural network predicted bitcoin prices are. His tool has only existed only a few weeks. It's only going to get better and more accurate. Scary accurate, I believe.

[K128kevin]

and skynet is born !

C'mon people, if it's right it's right, if it's wrong it's wrong... ether way just have fun.

That's the attitude xD

Kevin,

Thank you for the wonderful tool you have given us. I just hope you will continue to do so.

At this point in time, I think it's almost a moot point how accurate kevin's neural network predicted bitcoin prices are. His tool has only existed only a few weeks. It's only going to get better and more accurate. Scary accurate, I believe.

Glad you like it! And yeah I think the fact that it gets more and more accurate as time goes on is one of the coolest things about a project like this. There really is a relatively small amount of data available since bitcoin is so young, but as more data becomes available the accuracy should improve more and more!

[twiifm]

Describe what you are doing and ill tell you if human is better or not.   HFT does not predict.   HFT front run orders and capture the spread.   It can do this better humans cause it can do it faster.   

But humans program it to do this.

I dont know if a machine can find patterns better than humans.    Give me a real world example

Let's do a test.   We guess the open close price each day of S&P for next 10 days.  If you're right you get a point.   Whoever scores the most point wins.   Just an experiment to see if your algo can guess the price better than me.   

At market close you have to guess the open price.   At open you guess the closing price.  Your allowed to be off by +/- 50c

I'm sorry but I'm not going to bother trying to get my software to predict S&P right now because I have a ton of other work to do. If you want to test accuracy, you should just make your own bitcoin price predictions and test their accuracy compared to my charts on this website.

If you want to know how it works look up artificial neural networks on Google. This isn't some simple process that I can explain to you in one post. It's a very complex computational model that people have been developing and researching since the 1940s. You could teach multiple college courses just on neural networks alone. To be honest, it seems like you are refuting 70 years worth of research based on your gut feeling that humans would be better at this - and if we're going to be realistic, computers are better than humans at nearly all tasks involving any type of computation. They're surpassing us quickly in other realms as well. They have been able to beat us in chess for a long time now, they can beat us in almost any other game, and they can even drive cars better than us now. Is it so hard to believe that they can predict stock prices better than us?

Thats not what I said at all.  I said markets are random so its impossible to predict them. They are by nature unpredictable.   If you think you can predict future prices youd be a billionaire by now. Just put all capital into options and you can double it every day

I was just curious if your software was a good trader.   Thats why I offered the challenge.   Maybe the chaellenge should be human and computer starting w same amount of money and see who makes more after a certain period.

Its a cop out if you can't explain what you are doing to predict these prices.  Telling me to google neural networks doesn't make you know future prices more than me.   I admit don't know future prices.   

[K128kevin]

I'm sorry but I'm not going to bother trying to get my software to predict S&P right now because I have a ton of other work to do. If you want to test accuracy, you should just make your own bitcoin price predictions and test their accuracy compared to my charts on this website.

If you want to know how it works look up artificial neural networks on Google. This isn't some simple process that I can explain to you in one post. It's a very complex computational model that people have been developing and researching since the 1940s. You could teach multiple college courses just on neural networks alone. To be honest, it seems like you are refuting 70 years worth of research based on your gut feeling that humans would be better at this - and if we're going to be realistic, computers are better than humans at nearly all tasks involving any type of computation. They're surpassing us quickly in other realms as well. They have been able to beat us in chess for a long time now, they can beat us in almost any other game, and they can even drive cars better than us now. Is it so hard to believe that they can predict stock prices better than us?

Thats not what I said at all.  I said markets are random so its impossible to predict them. They are by nature unpredictable.   If you think you can predict future prices youd be a billionaire by now. Just put all capital into options and you can double it every day

I was just curious if your software was a good trader.   Thats why I offered the challenge.   Maybe the chaellenge should be human and computer starting w same amount of money and see who makes more after a certain period.

Its a cop out if you can't explain what you are doing to predict these prices.  Telling me to google neural networks doesn't make you know future prices more than me.   I admit don't know future prices.   
[/quote]

To say that something is, by nature, unpredictable is to say that it is random, which is literally the same as believing in magic. Nothing in the world is random - everything has an observable cause. If something has a cause, then it is not random. If it doesn't, then it is magic. That's what magic means. Maybe we don't have the means or technology to observe something, but it is still theoretically observable. This is besides the point though.

People VASTLY overestimate the power that comes with the ability to accurately predict bitcoin prices MOST of the time. This does not mean you will instantly become rich, or even become rich ever. It means that there is a higher percent chance for you to make some money on the bitcoin market. To think that someone could become a billionaire as an undergrad in college speculating on bitcoin prices is absolutely ridiculous. This isn't a magical get-rich-quick tool and I've expressed this repeatedly to everyone.

To say that I can't explain it in a single post is not a cop out. If you really want to understand it you have to do some research. There are a lot of things in life that can't be explained in a paragraph, and this is one of them. Also it's not my job to explain this to you. I have tons of other things I need to be working on. Asking me to explain to you how neural networks work would be like asking a chemistry professor how chemistry works. Sure, I could give you some brief, vague, high level answer but you still wouldn't fully understand how predictions are made.

[twiifm]

Markets are random. 

Read "Fooled By Randomness" Nassim Taleb.   Most of professional traders know this

I know how neural networks.   I wanna how YOU come up w these price predictions.

If you think you're so accurate why bitcoins?   Why not study S&P and trade the options on them?

Then move on to SPX  futures?   I know of one lady who turned like $300K into $40M in 3 years. She's a retail trader

[aminorex]

Markets are random. 

Read "Fooled By Randomness" Nassim Taleb.   Most of professional traders know this

Naïve poster, school thyself.  Random does not mean a coin toss.
In e.g. Taleb's usage it means that the most effective models we know are stochastic process models.  They make predictions. That is why we call them effective.

[K128kevin]

Markets are random. 

Read "Fooled By Randomness" Nassim Taleb.   Most of professional traders know this

I know how neural networks.   I wanna how YOU come up w these price predictions.

If you think you're so accurate why bitcoins?   Why not study S&P and trade the options on them?

Then move on to SPX  futures?   I know of one lady who turned like $300K into $40M in 3 years. She's a retail trader

Once again, saying that it is random is literally the same as saying it is magic. And I do mean literally.

I'll give you a brief explanation of how it works, but I doubt this will help:

I use a 3-layer neural network that trains on 3 years worth of historic data from bitstamp (about 4 million transactions I think). It looks at 60 inputs (average hourly prices in a row) and spits out 24 outputs (next 24 hourly averages). It trains about 30-35 times by making predictions at a few tens of thousands of points throughout the historic data and adjusting the weights of the edges between nodes according to the errors of its predictions. After about 30-35 times training on that set, it can predict 24 prices with an average error of about 1.3%.

And you have to realize, I don't know a lot about finance - I know a lot about computer science and data. I don't own any stocks and I don't have enough money to buy any stocks. I own 0.6 bitcoins and I can't afford to buy any more, so I'm just holding them. You need to have money to spare in order to try to make money on the bitcoin or stock market. I don't have that. It's not easy to make money in these markets unless you have plenty of extra money to spare, and even then I doubt it's easy.

[SportsBet]

Kevin,

Thank you for the wonderful tool you have given us. I just hope you will continue to do so.

At this point in time, I think it's almost a moot point how accurate kevin's neural network predicted bitcoin prices are. His tool has only existed only a few weeks. It's only going to get better and more accurate. Scary accurate, I believe.

and what kind of data is this belief based upon?|

@Kevin, I dont want to put you down, infact Id support you. But predictions have been consistently wrong and then adjusted to whats going on.

[comp006]

I think the main problem is that the more people use this chart the more it will be wrong because it can't take its own influence to the market into calculation. This is the main problem with EVERY public prediction.

[K128kevin]

and what kind of data is this belief based upon?|

@Kevin, I dont want to put you down, infact Id support you. But predictions have been consistently wrong and then adjusted to whats going on.

By adjusting the predictions every hour, accuracy is greatly increased. It allows itself to change its prediction if something happens that changes the course of prices. I know it can seem like it's changing its mind, but it rarely significantly changes the prediction.

I'm not sure what chart you've been looking at, but the 24 hour chart has been right much more often than it has been wrong over the past 3 weeks. The 5 day hasn't done quite as well but it has done decently. The 20 day one is not reliable, as I've explained before. I've started collecting data to prove this, and you'll be able to see charts with predicted prices compared against actual prices. This will be up within a week hopefully.

I think the main problem is that the more people use this chart the more it will be wrong because it can't take its own influence to the market into calculation. This is the main problem with EVERY public prediction.

That is a potential problem, though I don't think it is one yet. I don't think my site is popular enough yet to have a significant influence over the market. If/when it does become popular enough, I expect that changes in price might be more dramatic than it anticipates, and they might occur sooner as well. It definitely shouldn't predict the wrong direction of movement or anything though.

[twiifm]

Markets are random.  

Read "Fooled By Randomness" Nassim Taleb.   Most of professional traders know this

Naïve poster, school thyself.  Random does not mean a coin toss.
In e.g. Taleb's usage it means that the most effective models we know are stochastic process models.  They make predictions. That is why we call them effective.

No what we think are patterns is actually randomness.   Why are stochastic processes counter to what I say? All I claim is markets can't be predicted.  

Theres a difference in saying BTC might be $500 in 24hrs than BTC will be in 24hrs.

[K128kevin]

No what we think are patterns is actually randomness.   Thats why you trade probality not patterns.   Why are stochastic processes counter to what I say? All I claim is markets can't be predicted.  

Anyone who claims they can predict the future is either lucky or delusional

Have you been reading what I wrote at all? Do you still not understand that there is no randomness? This isn't an opinion, it's a fact. NOTHING IS RANDOM. EVER. Not even the random number generators in computers - it's based off of complicated strings and your cpu clock.

Do you realize that if you can predict whether the price of something is going to move up or down correctly more than 50% of the time, you are successfully predicting future prices? Nobody said they can predict prices perfectly 100% of the time.

I don't understand how you don't understand this -_- It just plain is NOT random. All I'm claiming is that I can predict the price within a reasonable margin of error MOST of the time. And it's not like you can argue against this, because it's objectively true. My software does it on the ENTIRE HISTORY OF BITCOIN DATA every hour and predicts within an average of 1.3% error. You're arguing against something that is just clearly objectively true.

Edit: Read the name of the post!!! I said it's a prediction, that that it WILL be $500! ugh please try to understand...

[twiifm]

No what we think are patterns is actually randomness.   Thats why you trade probality not patterns.   Why are stochastic processes counter to what I say? All I claim is markets can't be predicted.  

Anyone who claims they can predict the future is either lucky or delusional

Have you been reading what I wrote at all? Do you still not understand that there is no randomness? This isn't an opinion, it's a fact. NOTHING IS RANDOM. EVER. Not even the random number generators in computers - it's based off of complicated strings and your cpu clock.

Do you realize that if you can predict whether the price of something is going to move up or down correctly more than 50% of the time, you are successfully predicting future prices? Nobody said they can predict prices perfectly 100% of the time.

I don't understand how you don't understand this -_- It just plain is NOT random. All I'm claiming is that I can predict the price within a reasonable margin of error MOST of the time. And it's not like you can argue against this, because it's objectively true. My software does it on the ENTIRE HISTORY OF BITCOIN DATA every hour and predicts within an average of 1.3% error. You're arguing against something that is just clearly objectively true.

Edit: Read the name of the post!!! I said it's a prediction, that that it WILL be $500! ugh please try to understand...

Maybe you are lucky.   Nothing is random ever?   If I flip a coin tell me if heads or tails.   Put that in your neural network. And predict heads vs tails

Dude,  it wasnt 500.  Not even close.   Your prediction was wrong.   That is the empirical truth



  

[MysticalPotato]

Human behavior is not totally unpredictable. Especially given large numbers of humans.

I got chills, man.
That sounded like Hari Seldon talking there.

[K128kevin]

Maybe you are lucky.   Nothing is random ever?   If I flip a coin tell me if heads or tails.   Put that in your neural network. And predict heads vs tails

Dude,  it wasnt 500.  Not even close.   Your prediction was wrong.   That is the empirical truth

The outcome of a coin flip is determined by the force with which you hit the coin, how far from the center you hit it, and tons of other physical properties that are obviously extremely difficult to measure. This is why it effectively is random for us, but it's not truly random, as you claim bitcoin prices to be. Theoretically if you could observe every physical factor that would affect the outcome of a coin flip, you could know with 100% certainty what the outcome would be. This means that it is not random.

The coin flip is a terrible analogy to bitcoin prices. Coinflips have a nearly even distribution of heads vs tails (it's not actually even). Bitcoin price movement is not binary, and it is not an even distribution. A coin flip would be a valid analogy to a market where the value of the commodity is ALWAYS either $1 or $2, and NEVER anything else. And also where it is $1 50% of the time, and $2 50% of the time. I hope this shows you how flipping a coin obviously doesn't even come close to making sense as an analogy to bitcoin prices.

And as I explained to you earlier but maybe you forgot or didn't read - it readjusted its prediction one hour after I made this post, and it was pretty close after readjusting. It accurately predicted the rise in price, just not to the correct degree. I've also repeatedly said it's not ALWAYS right. Saying that one prediction is wrong (which it wasn't) and therefore none are right is like saying seeing one person win money at a casino and saying that casino NEVER earns money.

I got chills, man.
That sounded like Hari Seldon talking there.

I looked up who that is - that book looks kind of awesome! I might check it out.

[Beans]

I'm sorry but I'm not going to bother trying to get my software to predict S&P right now because I have a ton of other work to do. If you want to test accuracy, you should just make your own bitcoin price predictions and test their accuracy compared to my charts on this website.

If you want to know how it works look up artificial neural networks on Google. This isn't some simple process that I can explain to you in one post. It's a very complex computational model that people have been developing and researching since the 1940s. You could teach multiple college courses just on neural networks alone. To be honest, it seems like you are refuting 70 years worth of research based on your gut feeling that humans would be better at this - and if we're going to be realistic, computers are better than humans at nearly all tasks involving any type of computation. They're surpassing us quickly in other realms as well. They have been able to beat us in chess for a long time now, they can beat us in almost any other game, and they can even drive cars better than us now. Is it so hard to believe that they can predict stock prices better than us?

Thats not what I said at all.  I said markets are random so its impossible to predict them. They are by nature unpredictable.   If you think you can predict future prices youd be a billionaire by now. Just put all capital into options and you can double it every day

I was just curious if your software was a good trader.   Thats why I offered the challenge.   Maybe the chaellenge should be human and computer starting w same amount of money and see who makes more after a certain period.

Its a cop out if you can't explain what you are doing to predict these prices.  Telling me to google neural networks doesn't make you know future prices more than me.   I admit don't know future prices.  

To say that something is, by nature, unpredictable is to say that it is random, which is literally the same as believing in magic. Nothing in the world is random - everything has an observable cause. If something has a cause, then it is not random. If it doesn't, then it is magic. That's what magic means. Maybe we don't have the means or technology to observe something, but it is still theoretically observable. This is besides the point though.

People VASTLY overestimate the power that comes with the ability to accurately predict bitcoin prices MOST of the time. This does not mean you will instantly become rich, or even become rich ever. It means that there is a higher percent chance for you to make some money on the bitcoin market. To think that someone could become a billionaire as an undergrad in college speculating on bitcoin prices is absolutely ridiculous. This isn't a magical get-rich-quick tool and I've expressed this repeatedly to everyone.

To say that I can't explain it in a single post is not a cop out. If you really want to understand it you have to do some research. There are a lot of things in life that can't be explained in a paragraph, and this is one of them. Also it's not my job to explain this to you. I have tons of other things I need to be working on. Asking me to explain to you how neural networks work would be like asking a chemistry professor how chemistry works. Sure, I could give you some brief, vague, high level answer but you still wouldn't fully understand how predictions are made.

Quantum mechanics tells us that things are random. Even if you could chart every particle in the universe, you could not predict everything. Although, you can find likely outcomes. When it comes to predicting the price of bitcoin, it doesn't matter anyway. There are way to many unknown variables.

[K128kevin]

Quantum mechanics tells us that things are random. Even if you could chart every particle in the universe, you could not predict everything. Although, you can find likely outcomes.

Haha I was waiting for someone to bring up quantum mechanics. I feel like that's the only legitimate argument you can make against what I said. I don't understand enough about quantum mechanics to really discuss this, but from what I've heard from people who actually study it and know a decent amount about it (nobody knows "a lot" about quantum mechanics) all of the things about randomness are still pretty theoretical or just not well understood.

It's a very interesting subject, but regardless it doesn't conflict with my point that bitcoin prices are NOT random.

[Beans]

Quantum mechanics tells us that things are random. Even if you could chart every particle in the universe, you could not predict everything. Although, you can find likely outcomes.

Haha I was waiting for someone to bring up quantum mechanics. I feel like that's the only legitimate argument you can make against what I said. I don't understand enough about quantum mechanics to really discuss this, but from what I've heard from people who actually study it and know a decent amount about it (nobody knows "a lot" about quantum mechanics) all of the things about randomness are still pretty theoretical or just not well understood.

It's a very interesting subject, but regardless it doesn't conflict with my point that bitcoin prices are NOT random.

You should check out "Through the Wormhole", they spend a lot of time on the subject. Basically if you shoot a electron through a solid flat surface it would leave more then one hole. Unless you were observing it. May as well be magic, to our generation at least. We might just be living inside the matrix, that's just a glitch in the program..

[SportsBet]

Quantum mechanics tells us that things are random. Even if you could chart every particle in the universe, you could not predict everything. Although, you can find likely outcomes.

Haha I was waiting for someone to bring up quantum mechanics. I feel like that's the only legitimate argument you can make against what I said. I don't understand enough about quantum mechanics to really discuss this, but from what I've heard from people who actually study it and know a decent amount about it (nobody knows "a lot" about quantum mechanics) all of the things about randomness are still pretty theoretical or just not well understood.

It's a very interesting subject, but regardless it doesn't conflict with my point that bitcoin prices are NOT random.

You should check out "Through the Wormhole", they spend a lot of time on the subject. Basically if you shoot a electron through a solid flat surface it would leave more then one hole. Unless you were observing it. May as well be magic, to our generation at least. We might just be living inside the matrix, that's just a glitch in the program..

Learning about quantum mechanics from "Trough the wormhole" is the last thing I would advice.

[Dr Bloggood]

Interesting thread! OP, don't let yourself be thrown off the path by a couple of quite ignorant posters!

What are the medium and long term predictions? 5 months? 1 year? 3 years?

[toxic1978]

Great Thread. Nothing is Random, randomization is of course impossible as every action has a predetermined consequence. Whether the human mind or tools which the human mind has created have the processing ability to discover the results of all actions before they occur is of course unlikely at our current technological stage.

But having said that the Financial Markets are not as complex as say human life, therefor we are likely to be able to determine fairly accurate predictions of financial markets with the right input data, what of course makes these predictions less stable are variables which we do not have access to, such as other peoples influence. But again this is not random, we just don't have all the data we need.

[K128kevin]

Dr Bloggood - 20 days is the furthest I've been able to get the neural network to reasonably predict into the future, and unfortunately that one isn't actually especially accurate. The further forward you look, the more difficult it is to make predictions because real-world events that software can't anticipate will cause price changes.

toxic1978 - Yup! That's pretty much exactly what I'm trying to say.

catfish - That seems very interesting! So if I understand correctly, your idea was to have a neural network that could predict the behavior of individual traders in response to market changes, and then based on those predictions you could extrapolate to see what would happen to prices?

Also did they have multi-layered neural networks back when you were thinking of doing this? Because I know 3+ layers is relatively new and this would definitely be impossible with anything less than 3 layers. I guess computer speed back then might have made it difficult as well.

[Beans]

Great Thread. Nothing is Random, randomization is of course impossible as every action has a predetermined consequence. Whether the human mind or tools which the human mind has created have the processing ability to discover the results of all actions before they occur is of course unlikely at our current technological stage.

But having said that the Financial Markets are not as complex as say human life, therefor we are likely to be able to determine fairly accurate predictions of financial markets with the right input data, what of course makes these predictions less stable are variables which we do not have access to, such as other peoples influence. But again this is not random, we just don't have all the data we need.

It is by no means certain that actions have predetermined consequences, in fact it seems it's just not the case. If your right though, we had better stop wasting our time trying to build quantum computers.

[bitsmichel]

the 20 day prediction is looking good. how accurate is this?

which neural network did you implement? how much data did you fed it?

[K128kevin]

the 20 day prediction is looking good. how accurate is this?

which neural network did you implement? how much data did you fed it?

Be careful with the 20 day prediction, because it does have a large average error (about 8-8.2%). That one is not particularly reliable compared to the others, but I think it may be indicative of long-term trends. It has been much less than 20 days since I've had it running so I actually don't really know yet. We'll see I guess. The 24 hour prediction is pretty accurate most of the time though, and the 5 day one isn't quite as good but it still does reasonably well.

I implemented a 3-layer neural network that uses backpropogation. It trains on the entire history of transactions on bitstamp, so about 3 years worth of data.

[paulsonnumismatics]

Also honestly I find it hard to believe that anybody can believe that THIS is impossible. After all that humanity has done, all of our complex and great technological achievements, you think that predicting stock prices with reasonable accuracy is impossible? It's very possible and I have done it. So have many others.

You are claiming to create the first steps of PsicoHistory, my friend. If so, as an Asimov hardcore fan i welcome you. Let me be just skeptic, ok?

[Dotto]

Computer randomness is pretty pseudorandom, maked up with the ticks of the computers clock to amplify the randomness, but quantic random generators are fucking random itselves.

However when a massive number of humans focus on the same event (11S attack, new years eve, etc). The different quantic rnd generators display a very curious behaviour: they enter in synchronicity offering the same 0's and 1´s at the same time.

Strange.

It seems conscience has a major impact bending probabilities. The wave function collapse is another example in the same line, I bet that these facts are somehow related, but we don´t know how, yet.

Even if I have some reserves about the possibility of predicting BTC prices or tendencies with neural nets, I find your work fascinanting and will keep an eye on it.

[TwinWinNerD]

I like the website and the idea behind it, but it is not very accurate lately

I made the test and save a screenshot of the 24h graph every 24 hours. There is SOME correlation, but most swings where not accounted, or the opposite direction was predicted.

[K128kevin]

TwinWinNerD- Really? I've found that (with the exception of when the China ban thing happened) it almost never predicts the wrong direction of movement. It might predict the wrong direction if it, for example, says the price will go up by $2 but it ends up going down by $2 instead, which would be less than 1% error. For pretty much all of the major price changes I've watched it predict against, it's gotten the direction of movement correct but often overestimates the magnitude. There has been one or two instances where it predicted for the price to move up or down and it pretty much stayed the same, but I have not seen it predict a price change of more than like 2% where the actual price ended up moving in the opposite direction (except right after the China ban).

[Bit_Happy]

I predict $500 before $400 but not within 24 hours.
Slow and steady.........for now.

[K128kevin]

I predict $500 before $400 but not within 24 hours.
Slow and steady.........for now.

Predictions right now are showing a jump from mid 400s to just over 500 around the afternoon (EST) on 4/8. It's been consistently predicting this jump up for a while so I think we'll definitely at least see some kind of activity even if it's not right. I think we'll see a rise in price.

[BitOnyx]

You predict this prices or you just guess ?

[morphtrust]

Human behavior is not totally unpredictable. Especially given large numbers of humans.

I got chills, man.
That sounded like Hari Seldon talking there.

lol that is exactly what I was thinking too when I first saw this thread and what he was trying to do XD

[morphtrust]

Quantum mechanics tells us that things are random. Even if you could chart every particle in the universe, you could not predict everything. Although, you can find likely outcomes.

Haha I was waiting for someone to bring up quantum mechanics. I feel like that's the only legitimate argument you can make against what I said. I don't understand enough about quantum mechanics to really discuss this, but from what I've heard from people who actually study it and know a decent amount about it (nobody knows "a lot" about quantum mechanics) all of the things about randomness are still pretty theoretical or just not well understood.

It's a very interesting subject, but regardless it doesn't conflict with my point that bitcoin prices are NOT random.

You should check out "Through the Wormhole", they spend a lot of time on the subject. Basically if you shoot a electron through a solid flat surface it would leave more then one hole. Unless you were observing it. May as well be magic, to our generation at least. We might just be living inside the matrix, that's just a glitch in the program..

um what? pretty sure the original experiment was that they had two slits and they shot them through each causing a wave patter to appear much like spectrum lines from light passing through slits causing "banding" and when "observed" they stopped behaving like a "wave" and went back to behaving like a "particle" and just made two "piles" on either side, much the way sand pouring through two holes would cause two piles of sand on the other side,

I can tell you it is not "observation" that causes this, and the secret to figuring out how this oddity is occurring you just need to know HOW they were "observing" the action, since it was not with a human eye, and no human ever had to look at what was recorded to effect the outcome that can be seen as an after effect with out looking at it. the effect on the outcome was the same either way lol. it amazes me how little people really understand about things and how they hear chinese whispers about things and then just take it all literally, instead of going to the source and looking hard at the data.


the market is not random, that is for sure, there are events that cause things to occur, but knowing what to expect is going to take more than looking at the numbers each time, you need to look at spikes and then look at real events that tailor peoples lives, and see if there is a correlation, then see if the event has a pattern (like the fourth of july that happens here in the USA as a holiday where people blow lots of money on food fireworks and partying, to realize that people may cash out to do so, causing a crash, and then afterwards the market can repair itself as people go back to having just as much burden on their wallets as they did before things happened, sudden windfalls that happen say yearly will have significant effect on things, but not always, so you have to calculate all new changes and how they effect people negatively or positively, and your neural network at this point I am gathering can not do that, but when it can, then we will not see an unstable market anymore, unless you with hold that and use it to make millions lol.

[morphtrust]

TwinWinNerD- Really? I've found that (with the exception of when the China ban thing happened) it almost never predicts the wrong direction of movement. It might predict the wrong direction if it, for example, says the price will go up by $2 but it ends up going down by $2 instead, which would be less than 1% error. For pretty much all of the major price changes I've watched it predict against, it's gotten the direction of movement correct but often overestimates the magnitude. There has been one or two instances where it predicted for the price to move up or down and it pretty much stayed the same, but I have not seen it predict a price change of more than like 2% where the actual price ended up moving in the opposite direction (except right after the China ban).

two things here, one, the market numbers you are looking at, people speculating are also looking at, and I think are expecting things to happen so they are reducing the effect, by acting on this information, I have noticed that the market seems to swing back and forth between strong change and weak ones, and figure this is because people can see about a day or two back on the charts for the exchange they are on, so they are trying to play on that and cause it to not work,

two: you might ask the more important question, since it seems that it can change differently depending on the market you are on at the moment lol, so what market was twinwinerd looking at, and what all market data are you churning with your program?

[morphtrust]

Computer randomness is pretty pseudorandom, maked up with the ticks of the computers clock to amplify the randomness, but quantic random generators are fucking random itselves.

However when a massive number of humans focus on the same event (11S attack, new years eve, etc). The different quantic rnd generators display a very curious behaviour: they enter in synchronicity offering the same 0's and 1´s at the same time.

Strange.

It seems conscience has a major impact bending probabilities. The wave function collapse is another example in the same line, I bet that these facts are somehow related, but we don´t know how, yet.

Even if I have some reserves about the possibility of predicting BTC prices or tendencies with neural nets, I find your work fascinanting and will keep an eye on it.

ok one last post here lol. this sounds like you have been following the PEAR group at Princeton Uni. heh heh,

for those wondering what he might be talking about, you can check out www.psyleron.com I believe and see the REG (*random event generators) they have built into USB devices to record the changes, or the lights that you "think on/off" with your mind alone. I still have the software that was meant to train a person to use this "conscience" to their advantage, and did quite well against others in competitive mode (name of the software was shapechanger, and was put out by mindsong inc. ) 

[Predatorian]

Some of you have seen my website which tries to predict bitcoin prices with software using neural networks:

www.btcpredictions.com

It successfully predicted the drop in price that happened yesterday morning (EST) and it was very accurate throughout the day. It also predicted this rise that we are seeing now, and it's saying that prices will go into the upper $400s, possibly hitting $500! Here is an image of the predictions as of 11am EST:


The software is not always this accurate, but it has been performing exceptionally well recently! I think that we will see prices over $500 within the next 24 hours.

i hope u gonna be right with your ''prediction'' but imho we have to wait some more time for better price.

[johnytelevision]

Does this method make any sense ? From what I can see it look more like guessing then analyzing.

[Xell]

Predicting prices is of course possible, but you need to be clearer about what you mean when you say you are x% accurate.

Randomness is a concept, just like the number 3 or the imaginary number i^2=-1. You can't ever really observe it. People can talk forever about quantum physics but just because something appears random, doesn't mean it really is (in fact its probably not). It's only a lack of knowledge that makes us model these events as truly random (in my opinion, anyway). Even if something is truly random we can still make predictions, like I can bet that there will be less than seven heads if you toss a fair (ie, really random, 50/50) coin ten times. I can be more than 50% confident that will happen.

But that semantic aside, the real question is: how is average error measured? What do you even mean by 'error'? What model are you using for the underlying?

You can only be x% confident of an event happening if you know the underlying distribution (which you don't), or you are comparing to a model (which you haven't disclosed, and which is also obviously not totally correct).

Your graphs make no real sense, because there the probability of any of your graphs being correct is almost zero. What you (might) mean, is that based on your model you are 95% confident that the bitcoin price will stay within 1.273% of the line you predict. You can't simply have one error value and claim to be 100% confident that it will lie within that error. That's totally absurd.

Generally these things are modelled using something like Brownian motion (which is obviously not accurate, but is nice because it comes under many different process, Gaussian, Markov, Levy, Martingale etc., so there are many results we can use). But you simply can't have only one error value without also specifying a confidence level. You can model the BTC/USD price and claim based on your calculations you are 90% confident that the price will be within 450 and 480 in one hours time. But you can't claim 'average error' as a measure of current prediction accuracy. In fact, without defining what you even mean by 'error' then it may as well read 'average error: 3 aubergines and a courgette'. That would be equally useful.

In fact, you probably consider the current bitcoin price as continuous (rather than a discrete number of US$ & cents), in which case the probability you are correct is zero almost surely.

[K128kevin]

morphtrust - the possibility of people trading based on this information, and therefore causing higher errors is, of course, inevitable. However, at this point in the site's life I tend to doubt that enough people are actually making decisions based on this data to significantly affect the market though. This is very hard to measure though - especially given that I'm far from an expert in economics or markets and I don't really know how many people are needed to affect the market to what degree.

Xell - I mean, I thought the average error was pretty self explanatory. If it has an average error of 1.3%, that means that on average, it is off by 1.3%. That's about it. If you take all of its predictions and compare them against the actual prices, you'll find that the average of all the errors is 1.3%. A few people have asked about this though, so I'll give a little bit more detail.

The neural network, once it is done training, does some "test runs" where it attempts to make predictions at a few tens of thousands of points on the historic data. It compares its predictions to the actual prices that occurred and calculates the errors. The average errors that show up on the charts on the home page are the averages of all of these errors from the test runs.

With neural networks, there is no data that you would get which corresponds with something like a confidence level. It tries to approximate a function, but there is no probability involved in this.

[TwinWinNerD]

Xell - I mean, I thought the average error was pretty self explanatory. If it has an average error of 1.3%, that means that on average, it is off by 1.3%. That's about it. If you take all of its predictions and compare them against the actual prices, you'll find that the average of all the errors is 1.3%. A few people have asked about this though, so I'll give a little bit more detail.

1.3% of what? Absolute value? Pricemovement between 2 datapoints? Over the whole period?

[Xell]

morphtrust - the possibility of people trading based on this information, and therefore causing higher errors is, of course, inevitable. However, at this point in the site's life I tend to doubt that enough people are actually making decisions based on this data to significantly affect the market though. This is very hard to measure though - especially given that I'm far from an expert in economics or markets and I don't really know how many people are needed to affect the market to what degree.

Xell - I mean, I thought the average error was pretty self explanatory. If it has an average error of 1.3%, that means that on average, it is off by 1.3%. That's about it. If you take all of its predictions and compare them against the actual prices, you'll find that the average of all the errors is 1.3%. A few people have asked about this though, so I'll give a little bit more detail.

The neural network, once it is done training, does some "test runs" where it attempts to make predictions at a few tens of thousands of points on the historic data. It compares its predictions to the actual prices that occurred and calculates the errors. The average errors that show up on the charts on the home page are the averages of all of these errors from the test runs.

With neural networks, there is no data that you would get which corresponds with something like a confidence level. It tries to approximate a function, but there is no probability involved in this.

Just reread your post. That method of expressing error is hopeless when your predictions update hourly. It just doesn't make sense to take an average of errors like that. You need to explain better exactly how you calculate the error when (in the 24 hour case) you are really making 24 predictions per hour. You look at the error from all 24 predictions, square them (or take absolute value) and add them up? So you're adding 24 positive values up?

Your errors are also meaningless in terms of current prediction, because the errors you list are purely historic and are independent of the current prediction.

[K128kevin]

Just reread your post. That method of expressing error is hopeless when your predictions update hourly. It just doesn't make sense to take an average of errors like that. You need to explain better exactly how you calculate the error when (in the 24 hour case) you are really making 24 predictions per hour. You look at the error from all 24 predictions, square them (or take absolute value) and add them up? So you're adding 24 positive values up?

Your errors are also meaningless in terms of current prediction, because the errors you list are purely historic and are independent of the current prediction.

I think you are making it a lot more complicated than it is. I really calculate the error in the most basic way possible. Also I don't know what you are talking about as far as squaring anything or taking absolute values.

If I predict the price to be $500, and it turns out to be $505, that's a 1% error. It was off by $5, which is 1% of the predicted price. So I just take all of the predictions and calculate those errors, and take the average. And by taking the average, I mean taking the average, as in the average of 2, 2, 3, and 5 is 2 + 2 + 3 + 5 = 12 / 4 = 3. I just add them all together and divide by the number of predictions that were made to figure out what the average error is.

Average error really just means exactly what it says - on average, predictions are off by that percent.

[TwinWinNerD]

Just reread your post. That method of expressing error is hopeless when your predictions update hourly. It just doesn't make sense to take an average of errors like that. You need to explain better exactly how you calculate the error when (in the 24 hour case) you are really making 24 predictions per hour. You look at the error from all 24 predictions, square them (or take absolute value) and add them up? So you're adding 24 positive values up?

Your errors are also meaningless in terms of current prediction, because the errors you list are purely historic and are independent of the current prediction.

I think you are making it a lot more complicated than it is. I really calculate the error in the most basic way possible. Also I don't know what you are talking about as far as squaring anything or taking absolute values.

If I predict the price to be $500, and it turns out to be $505, that's a 1% error. It was off by $5, which is 1% of the predicted price. So I just take all of the predictions and calculate those errors, and take the average. And by taking the average, I mean taking the average, as in the average of 2, 2, 3, and 5 is 2 + 2 + 3 + 5 = 12 / 4 = 3. I just add them all together and divide by the number of predictions that were made to figure out what the average error is.

Average error really just means exactly what it says - on average, predictions are off by that percent.

Then you need to make that clear. Because right now it appears that you mean 1.3% over 24 hours. But in effect it is this between datapoints. Over the 24 datapoints, it can be off by 30% or so...

[K128kevin]

Then you need to make that clear. Because right now it appears that you mean 1.3% over 24 hours. But in effect it is this between datapoints. Over the 24 datapoints, it can be off by 30% or so...

No lol... it will never be off by 30% unless something ridiculous happens. This is the average error of all of the predictions in that time frame, it's not aggregate or anything. If you look at all of those points, on average each one is off by around 1.3%. Like I said, I just calculate how far off each one is and take the average of all of them.

[twiifm]

Yes very misleading.  1.5% error sounds like out of 100 predictions only 1.5 are wrong and 98.5% are correct

[Xell]

Just reread your post. That method of expressing error is hopeless when your predictions update hourly. It just doesn't make sense to take an average of errors like that. You need to explain better exactly how you calculate the error when (in the 24 hour case) you are really making 24 predictions per hour. You look at the error from all 24 predictions, square them (or take absolute value) and add them up? So you're adding 24 positive values up?

Your errors are also meaningless in terms of current prediction, because the errors you list are purely historic and are independent of the current prediction.

I think you are making it a lot more complicated than it is. I really calculate the error in the most basic way possible. Also I don't know what you are talking about as far as squaring anything or taking absolute values.

If I predict the price to be $500, and it turns out to be $505, that's a 1% error. It was off by $5, which is 1% of the predicted price. So I just take all of the predictions and calculate those errors, and take the average. And by taking the average, I mean taking the average, as in the average of 2, 2, 3, and 5 is 2 + 2 + 3 + 5 = 12 / 4 = 3. I just add them all together and divide by the number of predictions that were made to figure out what the average error is.

Average error really just means exactly what it says - on average, predictions are off by that percent.

I think you're making it a lot more simple than it really is. You have 24 published predictions for every hourly price. So 24h ago it predicted the price now. 23 hour ago it predicted a different price for now. Is the error based on the 24 hour prediction or the 1h prediction or a combination of those and all the predictions inbetween

As for making sure the values are positive just means if one prediction is 505 and the other 495 while it turns out to be 500 the average error is not zero.

[twiifm]

What he is giving is the mean of the variance.  Not % of error.  So what he is saying is the variance of his prediction is +/- 1.5% from the mean.

For standard deviation you have to take square root or you might get negative number

[Xell]

What he is giving is the mean of the variance.  Not % of error.  So what he is saying is the variance of his prediction is +/- 1.5% from the mean.

For standard deviation you have to take square root or you might get negative number

No he isn't. The error is purely historic. It has nothing to do with the current prediction.

[twiifm]

What he is giving is the mean of the variance.  Not % of error.  So what he is saying is the variance of his prediction is +/- 1.5% from the mean.

For standard deviation you have to take square root or you might get negative number

No he isn't. The error is purely historic. It has nothing to do with the current prediction.

OK, that's what it sounds like he's saying.  In any case I agree w you that it's misleading

[morphtrust]

morphtrust - the possibility of people trading based on this information, and therefore causing higher errors is, of course, inevitable. However, at this point in the site's life I tend to doubt that enough people are actually making decisions based on this data to significantly affect the market though. This is very hard to measure though - especially given that I'm far from an expert in economics or markets and I don't really know how many people are needed to affect the market to what degree.

Xell - I mean, I thought the average error was pretty self explanatory. If it has an average error of 1.3%, that means that on average, it is off by 1.3%. That's about it. If you take all of its predictions and compare them against the actual prices, you'll find that the average of all the errors is 1.3%. A few people have asked about this though, so I'll give a little bit more detail.

The neural network, once it is done training, does some "test runs" where it attempts to make predictions at a few tens of thousands of points on the historic data. It compares its predictions to the actual prices that occurred and calculates the errors. The average errors that show up on the charts on the home page are the averages of all of these errors from the test runs.

With neural networks, there is no data that you would get which corresponds with something like a confidence level. It tries to approximate a function, but there is no probability involved in this.

I think you mis read what I said, the price variance is lower from one year to the next because people are looking at "CHARTS" not "YOUR CHARTS"  basically the same reasons that cause spikes and dips are still happening but they are not as severe because people trying to use that to make money and pressing the event they know BECAUSE IT HAPPENED BEFORE" (not because they even know about your site or have ever been on it) your site would be  possibly effecting that if there were enough people who know about it yes, but right now I doubt that is the case as well, I have not been able to figure out what market your chart is so accurate to, because I trade on BTC38 and it was not very accurate when I checked it, and that is fine because I doubt the people there are on the same pulses the people in general are on your chart, since their market often does not mirror what other places are doing ever,

[Costanza1]

Just wanted to say this is a very interesting project/thread and I'm enjoying the discussion.  I too agree the error % is confusing.  If you have 24 different predictions of the price at a specific time do you average all 24 of those predictions and then compare that average to the actual price to find your error?  As your 24hr predictions are more accurate than your 5 day predictions wouldnt that also mean that predictions for an hour ahead are far more accurate than predictions 23 hours ahead?  Have you calculated error based on x hours in advance as opposed to the whole 24 hour chart?

Thanks!

[K128kevin]

Xell - Of course the error is historic. I can't compare my predictions to future prices...

In your example of a prediction being $495 or $505 and the actual price being $500, those two options would be two different average errors. If you predict $505 and the price is 500, that is closer than if you predict $495 and it ends up being $500 because the difference ($5) is a smaller percent of $505 than it is of $495.

You were asking about whether the error was based on the 24 hour prediction or the 1 hr prediction or something in between - it is based on all of them. I still think you're making it more complicated than it is. I look at every predicted price, calculate how far off it is from the actual price at that time, and then take the average of those errors. So in reality if you were to average all of the predictions made 24 hours out, the error would probably be a bit higher. However, if you were to average the predictions that are 1 hour out, it would be lower. Really the best way to look at it though is just that the predicted prices (individually) are each off by an average of 1.3%.

morphtrust - It uses bitstamp. Are you saying that the patterns in price fluctuation are different now because people have seen the older patterns before? Because if so, I would definitely have to disagree. I'm still not exactly sure if that's what you are saying though.

Costanza1 - Yes, it averages all of the predictions' errors for each chart. The 24 hour chart averages the 1 hr, 2 hr, etc up to 24 hour errors. I have not calculated the error at individual hours in advance, but I imagine it would show lower errors for the shorter predictions and larger errors for the longer ones. I don't think the difference between the shorter and longer ones would be particularly large though.

----------

Maybe to some people the error is confusing, and I don't want that. It is definitely not misleading though. To say it is misleading implies that the error makes it seem like it's more accurate than it is, which is false. The error effectively describes the neural network's accuracy.

For those of you who find the average error confusing, I would be very open to your ideas on how to make it more intuitive. Given a set of predicted prices, how would you describe their accuracy? To me the most intuitive thing was just to calculate the error of each prediction and take the average of those errors, so that's what I've done. I just don't see a simpler, more obvious way to do it than that. But if you guys have ideas I would love to hear them.

[Xell]

Xell - Of course the error is historic. I can't compare my predictions to future prices...

In your example of a prediction being $495 or $505 and the actual price being $500, those two options would be two different average errors. If you predict $505 and the price is 500, that is closer than if you predict $495 and it ends up being $500 because the difference ($5) is a smaller percent of $505 than it is of $495.

You were asking about whether the error was based on the 24 hour prediction or the 1 hr prediction or something in between - it is based on all of them. I still think you're making it more complicated than it is. I look at every predicted price, calculate how far off it is from the actual price at that time, and then take the average of those errors. So in reality if you were to average all of the predictions made 24 hours out, the error would probably be a bit higher. However, if you were to average the predictions that are 1 hour out, it would be lower. Really the best way to look at it though is just that the predicted prices (individually) are each off by an average of 1.3%.

twiifm (and I originally) was suggesting you had a statistical model (maybe time-series based), in which case you can give confidence bounds on your predictions based on your model. I'm not a computer scientist, but I'd be surprised if this wasn't possible in your case too, rather than keeping your error purely historic.

Maybe to some people the error is confusing, and I don't want that. It is definitely not misleading though. To say it is misleading implies that the error makes it seem like it's more accurate than it is, which is false. The error effectively describes the neural network's accuracy.

For those of you who find the average error confusing, I would be very open to your ideas on how to make it more intuitive. Given a set of predicted prices, how would you describe their accuracy? To me the most intuitive thing was just to calculate the error of each prediction and take the average of those errors, so that's what I've done. I just don't see a simpler, more obvious way to do it than that. But if you guys have ideas I would love to hear them.

It is misleading, because you are quoting an average error for a 24h period - your error increases the further away from 'now' you are, and what you imply is that this is an error for the 24h period when in fact the error for now+24h is greater than what you state. It's also misleading because you state ``[the errors] never rise more than 0.1% above these numbers.'' so 1.4% for the 24h case. But as the title of this thread shows, clearly that's just a nonsense.

As for ideas, here is my idea for accuracy: Instead of one number for the 24h period, have an 'average error' for each predicted price (that is, an average error for prices predicted 1,2,3,4,...,24h into the future) and plot a band rather than a line to show them. You might also consider a toggle function to only consider errors since the start of this year. My reason for suggesting that is there has been a different trend since January, seems different from the overall trend since Bitcoin began.

Also, anyone else should note that that bitcoin prices have been show to correlate most strongly with media cycles, rather than historic price cycles. Of course, you can show anything if you try hard enough with poor statistical tests but it's worth bearing in mind when considering the predictive power of a model that doesn't take media coverage into consideration.

[K128kevin]

Xell - A confidence interval just wouldn't make sense with this model of prediction. It would be like if you said that you are 90% confident that the value of the function f(x) = x + 5 is 7 given 2 for x, and 10% sure that it is 8 or 6. So that's not an option. Also error calculation HAS to be historic. There is no other way it could ever possibly be done - you can't measure the error of a prediction that has yet to play itself out.

The 24 hour prediction is probably a small amount higher than average, and the 1 hour prediction is probably a small amount lower - yes, that is true. I could possibly somehow inform viewers of the error of each hour individually, but I am 100% positive that this would be much more confusing to the vast majority of people. It would provide a lot of information, cluttering the charts and very few people would understand it without having to ask for an explanation. The way it is now, the vast majority of the 13,000 unique people who have viewed it so far have not needed an explanation, and the few who I have explained it to understood it pretty quickly. I would really like for everyone to be able to understand it pretty easily, but from what you guys are saying here it seems like, unfortunately, that may not be possible.

Also I don't think anybody is assuming that the average error quoted above the 24 hour chart refers ONLY to the last prediction. I'm pretty positive of this lol...

I would be curious to see what study you are referring to that showed that bitcoin prices correlate more with media patterns than historic patterns, but regardless this is irrelevant. As I've explained before, the cause of a recurring pattern in price does not matter. As long as some pattern exists, it can be recognized and used to successfully predict future prices. It can be the media causing the pattern, group psychology, or flying spaghetti monsters. Neural networks don't need to know this information.

And remember, this is finding patterns that are WAAYYYY too complicated for any human to see or understand. Just because something looks like it is the case doesn't mean that it is. It may look like price movements are arbitrary and they don't repeat themselves, but that just means that you cannot see the underlying patterns because our brains are just incapable of this.

[Xell]

Xell - A confidence interval just wouldn't make sense with this model of prediction. It would be like if you said that you are 90% confident that the value of the function f(x) = x + 5 is 7 given 2 for x, and 10% sure that it is 8 or 6. So that's not an option. Also error calculation HAS to be historic. There is no other way it could ever possibly be done - you can't measure the error of a prediction that has yet to play itself out.

The 24 hour prediction is probably a small amount higher than average, and the 1 hour prediction is probably a small amount lower - yes, that is true. I could possibly somehow inform viewers of the error of each hour individually, but I am 100% positive that this would be much more confusing to the vast majority of people. It would provide a lot of information, cluttering the charts and very few people would understand it without having to ask for an explanation. The way it is now, the vast majority of the 13,000 unique people who have viewed it so far have not needed an explanation, and the few who I have explained it to understood it pretty quickly. I would really like for everyone to be able to understand it pretty easily, but from what you guys are saying here it seems like, unfortunately, that may not be possible.

Also I don't think anybody is assuming that the average error quoted above the 24 hour chart refers ONLY to the last prediction. I'm pretty positive of this lol...

I would be curious to see what study you are referring to that showed that bitcoin prices correlate more with media patterns than historic patterns, but regardless this is irrelevant. As I've explained before, the cause of a recurring pattern in price does not matter. As long as some pattern exists, it can be recognized and used to successfully predict future prices. It can be the media causing the pattern, group psychology, or flying spaghetti monsters. Neural networks don't need to know this information.

And remember, this is finding patterns that are WAAYYYY too complicated for any human to see or understand. Just because something looks like it is the case doesn't mean that it is. It may look like price movements are arbitrary and they don't repeat themselves, but that just means that you cannot see the underlying patterns because our brains are just incapable of this.

In your example, your confidence interval would be a measure of how sure you are the function really looks like that over the next 60mins, based on historic info.

I'm fed up of arguing about your definition of error and your inappropriate simplification and presenting only a mean. It's a bad measure of error for the data you are presenting, and an inappropriate statistic on it's own. As it stands, the data you present is meaningless to everyone but you. It doesn't matter how many people visit the page, if they don't have any mathematical background then they won't understand the flaw in your 'average error' statement.

Lastly, thanks for the caps, but I'm not a child. I never mentioned humans being able to 'see' patterns in datasets. But humans can 'understand' patterns, in many cases. Bitcoin prices make a timeseries, and there are many, many, many ways to find patterns in time series (& judge how likely they are to 'really' exist) and understand exactly what they mean.

[Beans]

Quantum mechanics tells us that things are random. Even if you could chart every particle in the universe, you could not predict everything. Although, you can find likely outcomes.

Haha I was waiting for someone to bring up quantum mechanics. I feel like that's the only legitimate argument you can make against what I said. I don't understand enough about quantum mechanics to really discuss this, but from what I've heard from people who actually study it and know a decent amount about it (nobody knows "a lot" about quantum mechanics) all of the things about randomness are still pretty theoretical or just not well understood.

It's a very interesting subject, but regardless it doesn't conflict with my point that bitcoin prices are NOT random.

You should check out "Through the Wormhole", they spend a lot of time on the subject. Basically if you shoot a electron through a solid flat surface it would leave more then one hole. Unless you were observing it. May as well be magic, to our generation at least. We might just be living inside the matrix, that's just a glitch in the program..

um what? pretty sure the original experiment was that they had two slits and they shot them through each causing a wave patter to appear much like spectrum lines from light passing through slits causing "banding" and when "observed" they stopped behaving like a "wave" and went back to behaving like a "particle" and just made two "piles" on either side, much the way sand pouring through two holes would cause two piles of sand on the other side,

I can tell you it is not "observation" that causes this, and the secret to figuring out how this oddity is occurring you just need to know HOW they were "observing" the action, since it was not with a human eye, and no human ever had to look at what was recorded to effect the outcome that can be seen as an after effect with out looking at it. the effect on the outcome was the same either way lol. it amazes me how little people really understand about things and how they hear chinese whispers about things and then just take it all literally, instead of going to the source and looking hard at the data.


the market is not random, that is for sure, there are events that cause things to occur, but knowing what to expect is going to take more than looking at the numbers each time, you need to look at spikes and then look at real events that tailor peoples lives, and see if there is a correlation, then see if the event has a pattern (like the fourth of july that happens here in the USA as a holiday where people blow lots of money on food fireworks and partying, to realize that people may cash out to do so, causing a crash, and then afterwards the market can repair itself as people go back to having just as much burden on their wallets as they did before things happened, sudden windfalls that happen say yearly will have significant effect on things, but not always, so you have to calculate all new changes and how they effect people negatively or positively, and your neural network at this point I am gathering can not do that, but when it can, then we will not see an unstable market anymore, unless you with hold that and use it to make millions lol.

It works basically the way I said it. They shoot a electron through two slits which gives a wave pattern because the electron goes through both slits at the same time. When they observe each slit to actually see it happen, it changes and only goes through one slit at a time. It has been tested with the detector on but not collecting data. If any data is recorded telling us one way or the other, the wave pattern disappears. Look up the double-slit experiment if you need a lesson. I was trying to keep it simple because it can be hard to understand. Clearly you don't get it, but I guess you already debunked quantum mechanics and without a single shred of evidence. Obviously looking at the data doesn't effect the test, it's the act of measurement that forces a particle to choose a specific location.

[K128kevin]

In your example, your confidence interval would be a measure of how sure you are the function really looks like that over the next 60mins, based on historic info.

I'm fed up of arguing about your definition of error and your inappropriate simplification and presenting only a mean. It's a bad measure of error for the data you are presenting, and an inappropriate statistic on it's own. As it stands, the data you present is meaningless to everyone but you. It doesn't matter how many people visit the page, if they don't have any mathematical background then they won't understand the flaw in your 'average error' statement.

Lastly, thanks for the caps, but I'm not a child. I never mentioned humans being able to 'see' patterns in datasets. But humans can 'understand' patterns, in many cases. Bitcoin prices make a timeseries, and there are many, many, many ways to find patterns in time series (& judge how likely they are to 'really' exist) and understand exactly what they mean.

The confidence interval would not be a measure of how sure I am the function looks like that because this data does not exist.

I just don't understand how you can argue that this way of measuring error is meaningless and then have absolutely no better way of doing it yourself. You seem to know nothing about neural networks and how they calculate predictions like this and your suggestions have been either impossible or just much worse and more confusing than what I have now. My point was just that the vast majority of people understand the error easily. It's not an oversimplification and it makes a lot of sense to almost everyone. It's far from meaningless, and you have yet to give any reason at all to back your claim other than the fact that there will be slight differences in accuracy between the shorter and longer term data points on the same chart (which is a very weak argument).

If you are going to come here and tell me the data produced by my project (which I've spent months and months and a huge amount of effort working on) is "meaningless" then you should have a better than tenuous argument as to why, and you should have some kind of better way that you would have done it. You seem to have neither of these things, and I would be much more inclined to take your advice if you did.

And I mentioned this before but trying to understand the functions that a neural network finds in data would be kind of like, in this case, trying to understand a function with 60 * 200 * 24 = 288,000 different variables. This is far from simple stuff. This is how computers recognize voices and words, it is how they identify images, it is how Google brain can differentiate between humans and cats, it is how stock market prices can be predicted, and there are dozens of other very complicated tasks that they perform. This is much more complicated than just finding some pattern in a timeseries. This is artificial intelligence and machine learning.

[Xell]

The confidence interval would not be a measure of how sure I am the function looks like that because this data does not exist. They're always used to show confidence for predictions of days that doesn't exist yet.

I just don't understand how you can argue that this way of measuring error is meaningless and then have absolutely no better way of doing it yourself. You seem to know nothing about neural networks and how they calculate predictions like this and your suggestions have been either impossible or just much worse and more confusing than what I have now. My point was just that the vast majority of people understand the error easily. It's not an oversimplification and it makes a lot of sense to almost everyone. It's far from meaningless, and you have yet to give any reason at all to back your claim other than the fact that there will be slight differences in accuracy between the shorter and longer term data points on the same chart (which is a very weak argument).

If you are going to come here and tell me the data produced by my project (which I've spent months and months and a huge amount of effort working on) is "meaningless" then you should have a better than tenuous argument as to why, and you should have some kind of better way that you would have done it. You seem to have neither of these things, and I would be much more inclined to take your advice if you did.

And I mentioned this before but trying to understand the functions that a neural network finds in data would be kind of like, in this case, trying to understand a function with 60 * 200 * 24 = 288,000 different variables. This is far from simple stuff. This is how computers recognize voices and words, it is how they identify images, it is how Google brain can differentiate between humans and cats, it is how stock market prices can be predicted, and there are dozens of other very complicated tasks that they perform. This is much more complicated than just finding some pattern in a timeseries. This is artificial intelligence and machine learning.

Confidence intervals don't work like that. If the data exists then you don't state a confidence interval. You never know the data when you state confidence intervals.

Here's why your error is a poor stat. One, your error makes a time series itself. Condensing it like that and keeping all the historic results in the current error is not useful. Your error tells the user very little about recent applicability. Secondly, you don't make any effort to tell the user how your mean varies. The minimum is zero and the mean is 1.3% okay, but how about the variance? What's the tail look like? That's a very important piece of information that anyone who is more than slightly curious will miss from your site. Listing a mean with no variance or other info is meaningless when the underlying distribution is largely unknown.

I know very little about the method used to generate your model. I know nothing about AI. I'm a mathematician I spend all my time modelling. You shouldn't condense 24 time series (error) into one number with equal weighting like that it isn't a relevant statistic to anyone but you. That means your predictions for the price 2 years ago are as relevant as your ability to predict yesterdays price as far as your error goes.

The vast majority don't understand the error at all. They just don't look at it long enough to understand it or simply don't care or know enough to see why its flawed.

PS there are confidence interval methods for neural network models. Your method for calculation is an innaproriate statistic, no matter how many people 'understand it'.

[bitcoinforhelp]

we will see soon if you were right, i personally doubt it

[K128kevin]

Confidence intervals don't work like that. If the data exists then you don't state a confidence interval. You never know the data when you state confidence intervals.

Here's why your error is a poor stat. One, your error makes a time series itself. Condensing it like that and keeping all the historic results in the current error is not useful. Your error tells the user very little about recent applicability. Secondly, you don't make any effort to tell the user how your mean varies. The minimum is zero and the mean is 1.3% okay, but how about the variance? What's the tail look like? That's a very important piece of information that anyone who is more than slightly curious will miss from your site. Listing a mean with no variance or other info is meaningless when the underlying distribution is largely unknown.

I know very little about the method used to generate your model. I know nothing about AI. I'm a mathematician I spend all my time modelling. You shouldn't condense 24 time series (error) into one number with equal weighting like that it isn't a relevant statistic to anyone but you. That means your predictions for the price 2 years ago are as relevant as your ability to predict yesterdays price as far as your error goes.

The vast majority don't understand the error at all. They just don't look at it long enough to understand it or simply don't care or know enough to see why its flawed.

PS there are confidence interval methods for neural network models. Your method for calculation is an innaproriate statistic, no matter how many people 'understand it'.

As someone who knows nothing about neural networks, I don't think you are in any place to say that my method of calculation is an inappropriate statistic (also that statement doesn't make sense in English).

As someone who knows much less about who does and doesn't understand what aspects of my website than I do, I don't think you are in any place to say that the vast majority of people don't understand the error at all (which is false).

As someone who knows very little, if anything, about pattern recognition and function approximation, I don't think you are in any place to say that the error calculated on data from 2 years ago is any less relevant than the error calculated on data from 2 days ago. They are equally relevant and the fact that you disagree with this just continues to confirm your lack of understanding of my method.

The only reasonable point you've made is that I could show the variance distribution. However, I address this on the about page in a very non-mathematical way. I tell people that the prices are less accurate when real-life events are affecting the price, and they are more accurate when prices are stable. Statistics are not needed to back this statement, as it is pretty intuitively obvious and it gives people enough information to make their own intelligent judgement without having to clutter the site with a bunch of statistics that are meaningless to almost everyone (except you and the 3 other mathematicians who are looking at my site).
 
There are people out there who are skeptical of my predictions and even skeptical of my error figures, and that is perfectly reasonable. However, it is ridiculous when people who have no understanding of what I do criticize my methods. I understand that you have some background in math, but you clearly don't understand how non-mathematicians (i.e. almost every person on Earth) look at things and understand data. You are the only person of 13,000 who has expressed any issue with my method of calculating this number. I could clutter the site with tons of detailed statistics and you might be able to have a thorough understanding of the error, but nobody else would.

My current average error is one number - simple. It describes all of the data at once, and it is not misleading because everyone knows the 1hr prediction is likely to be more accurate than the 24hr prediction.

You can continue to hate on my project for no reason if you want, but you have no point here, I'm sorry...

[Easy2Mine]

Even if you improve the preditions and add more functions into your website, there are always people arguing about it, instead of looking for solutions to improved it themself.
Most people who have seen your work are positive.

[twiifm]

Just curious how do you trade these predictions?  You just buy and sell if your target is hit?

If your target doesn't hit what do you do?  Double down? Swallow loss?

[K128kevin]

Even if you improve the preditions and add more functions into your website, there are always people arguing about it, instead of looking for solutions to improved it themself.
Most people who have seen your work are positive.

Thank you, that's nice to hear

Just curious how do you trade these predictions?  You just buy and sell if your target is hit?

If your target doesn't hit what do you do?  Double down? Swallow loss?

Are you talking about the trade simulator idea that I've been testing? The method I'm testing at the moment, which I think is the most promising, is that it looks at the 24 hour chart and sets a buy and sell price based on the predictions. If it doesn't predict for the price to significantly change, it just sets the buy price to 1.5% below the current price, and the sell price to 1.5% above. If the price is supposed to change soon, then it sets the buy or sell price to the price that the chart says it should reach (buy if it's low, sell if it's high obviously).

If it doesn't reach the target then it will just wait. I have it so that it only buys and sells 1 btc at a time, never more or less. I feel like this is the simplest way to show potential profitability. Then people can see how much money per bitcoin it makes by trading based on predictions.

Also remember that if the target says, for example, that the price will be $480 at 6:00 pm, that doesn't actually mean it will be the price at that time. It means that will be the average price over the course of the hour leading up to that time, meaning that it is likely that the price will go a bit higher and a bit lower. If I set the sell to $480, that might not be the highest price it gets to but it will most likely reach that price. Even if the average price over the course of that hour ends up being $475 or $470, there is still a solid chance it will hit $480 at some point.

Unfortunately, the price has been super stable the past day or so, so the simulator isn't buying or selling. Hopefully we'll get a bit of volatility soon and we can see how it actually does.

[twiifm]

I hope you make a lot of money.   Im not here to criticize your system.   Im just skeptical as an options trader. Im a firm believer in random walk in short term (minutes) ,  technicals in semi  short (hours,  days)  term and fundamentals long term (days,  months) .   Most of my trades are delta neutral.  I sell premium 45 days out on IV spikes.   I do have spec long positions that are 6 mos to year out

2 problems I foresee trading BTC

1.  The slippage on BTC  is huge.   Probably  cause theres no liquidity.  I looked on bitstamp and coinbase and the bid/ask can be $5 wide.  Lack of liquidity also makes difficult for intraday trading if you cant get in and out of position

2.  Theres no easy way to short BTC so you can only trade bullish direction.   In options you can trade up,  down,  or sideways

[K128kevin]

I hope you make a lot of money.   Im not here to criticize your system.   Im just skeptical as an options trader. Im a firm believer in random walk in short term (minutes) ,  technicals in semi  short (hours,  days)  term and fundamentals long term (days,  months) .   Most of my trades are delta neutral.  I sell premium 45 days out on IV spikes.   I do have spec long positions that are 6 mos to year out

2 problems I foresee trading BTC

1.  The slippage on BTC  is huge.   Probably  cause theres no liquidity.  I looked on bitstamp and coinbase and the bid/ask can be $5 wide.  Lack of liquidity also makes difficult for intraday trading if you cant get in and out of position

2.  Theres no easy way to short BTC so you can only trade bullish direction.   In options you can trade up,  down,  or sideways

Yeah, I mean as I've said, I don't know a lot about trading. I'm doing this from a computer science perspective, so I don't know how profitable this will be. As I've said before though, I believe predictions should be taken with a grain of salt, and you should consider recent news as well when making decisions, and use your own judgement. I would never blindly trade according to these predictions only (which is what the simulator will be doing). However I believe it can be a good tool to consult when making decisions to buy/sell. It may be able to help you optimize when you decide to execute trades, allowing you to earn some extra money.

[toxic1978]

Great Thread. Nothing is Random, randomization is of course impossible as every action has a predetermined consequence. Whether the human mind or tools which the human mind has created have the processing ability to discover the results of all actions before they occur is of course unlikely at our current technological stage.

But having said that the Financial Markets are not as complex as say human life, therefor we are likely to be able to determine fairly accurate predictions of financial markets with the right input data, what of course makes these predictions less stable are variables which we do not have access to, such as other peoples influence. But again this is not random, we just don't have all the data we need.

It is by no means certain that actions have predetermined consequences, in fact it seems it's just not the case. If your right though, we had better stop wasting our time trying to build quantum computers.

Find me a theory that explains how a random activity works. For something to be random it has to happen for no reason, with nothing preceeding it. We cannot as humans beings necessarily predict the consequences of actions, but how can something happen without cause or reason?

Did I randomly come across this post and decide to start writing a response.... well if I was telling someone else about it I would probably say I came across the post randomly - however I didn't, all the events from my birth onwards led me to that point (and of course my mother and fathers actions would of also had an affect, and theirs and so on). So it wasn't random, several trillion variables which all had a set value due to other trillions of variables and so on created the unique event of me replying to this post.

But hey, its just theory of mine - I may be wrong .

:p

[twiifm]

Great Thread. Nothing is Random, randomization is of course impossible as every action has a predetermined consequence. Whether the human mind or tools which the human mind has created have the processing ability to discover the results of all actions before they occur is of course unlikely at our current technological stage.

But having said that the Financial Markets are not as complex as say human life, therefor we are likely to be able to determine fairly accurate predictions of financial markets with the right input data, what of course makes these predictions less stable are variables which we do not have access to, such as other peoples influence. But again this is not random, we just don't have all the data we need.

It is by no means certain that actions have predetermined consequences, in fact it seems it's just not the case. If your right though, we had better stop wasting our time trying to build quantum computers.

Find me a theory that explains how a random activity works. For something to be random it has to happen for no reason, with nothing preceeding it. We cannot as humans beings necessarily predict the consequences of actions, but how can something happen without cause or reason?

Did I randomly come across this post and decide to start writing a response.... well if I was telling someone else about it I would probably say I came across the post randomly - however I didn't, all the events from my birth onwards led me to that point (and of course my mother and fathers actions would of also had an affect, and theirs and so on). So it wasn't random, several trillion variables which all had a set value due to other trillions of variables and so on created the unique event of me replying to this post.

But hey, its just theory of mine - I may be wrong .

:p

Random as in unpredictable.   

[Bit_Happy]

$500 within 24 hours from now is still possible.
BTC tends to bounce fast once the rally/recovery begins. 

[K128kevin]

$500 within 24 hours from now is still possible.
BTC tends to bounce fast once the rally/recovery begins. 

lol well I hope it goes up, but also I posted this a week ago haha

[LordMo]

Is the algorithm based on some sort of last couple of days moving average? It constantly says price will go up.

[alero]

$500 getting less and less probable...

Current price: $377.02 (14.35%)

[K128kevin]

Is the algorithm based on some sort of last couple of days moving average? It constantly says price will go up.

It doesn't always say that the price will go up, it has successfully predicted dips in price before.

But yes, it looks at the last 60 hours for the 24 hour prediction. However, it learns how to predict by looking at the entire price history on bitstamp, going about 3 years back. The model it uses is called an artificial neural network. You can read about them on Wikipedia or a bunch of other sites if you are interested.

[vinipoars]

Most probably:
350 within 1 week.
300 within 2 weeks.
250 within 3 weeks
200 within 1 month.

 

[K128kevin]

Most probably:
350 within 1 week.
300 within 2 weeks.
250 within 3 weeks
200 within 1 month.

 

Hmmm I think that the price will move faster than that. If it keeps crashing, I think it will be below $200 in one month. Otherwise, I think it will have spiked back up to at least like $500. I would be surprised to see a slow and steady fall like that.

[porcupine87]

Even if you improve the preditions and add more functions into your website, there are always people arguing about it, instead of looking for solutions to improved it themself.
Most people who have seen your work are positive.

I don't know if this method adds any value. You read about an error of 1.3% on a daily basis. This is a flaw because this only counts for ex ante predictions. Not for acutal predictions into the future. For example he made the prediction of 480$. The real price was around 440$. This is an error of 8.3%.

I watched it for a while and checked:
30.march 3pm it predicted 489$. Real price (24h later) was: 455$ (7%error)
31.3. 12am he predicted 465$. Real price was: 486$ (4.3%)
1.4. 1am he predicted 456$. Real price was: 485$ (6.3%)
1.4. 8pm he predicted 466$. Real price was: 434$ (6.9%)
2.4. 11am he predicted 449$. Real price was: 422$ (6%)

So there is an average error of about 6%.

So is that good or bad? We can just estimate that quickly. Take the prediction: "Price today is the price in 24h". This is just the volatility. I took the hourly prices of bitstamp and compared it to the value 24h later. The mean error was 3.8%.

-> So if you just guess that the current price is the price in 24h you drive better than with this prediction of the neural network (although I just made a sample of 6 predictions of it).

[K128kevin]

I don't know if this method adds any value. You read about an error of 1.3% on a daily basis. This is a flaw because this only counts for ex ante predictions. Not for acutal predictions into the future. For example he made the prediction of 480$. The real price was around 440$. This is an error of 8.3%.

I watched it for a while and checked:
30.march 3pm it predicted 489$. Real price (24h later) was: 455$ (7%error)
31.3. 12am he predicted 465$. Real price was: 486$ (4.3%)
1.4. 1am he predicted 456$. Real price was: 485$ (6.3%)
1.4. 8pm he predicted 466$. Real price was: 434$ (6.9%)
2.4. 11am he predicted 449$. Real price was: 422$ (6%)

So there is an average error of about 6%.

So is that good or bad? We can just estimate that quickly. Take the prediction: "Price today is the price in 24h". This is just the volatility. I took the hourly prices of bitstamp and compared it to the value 24h later. The mean error was 3.8%.

-> So if you just guess that the current price is the price in 24h you drive better than with this prediction of the neural network (although I just made a sample of 6 predictions of it).

There are several things wrong with this data you are providing.

First of all, I'm not sure how you chose the specific times you tested, but they are clearly it's worst predictions. As you said, you only took a sample of 6 predictions. Every hour I take a sample of something like 60,000 predictions and calculate the average error. It comes to about 1.3%, so actually that is the correct figure (not 6%). Furthermore, this means that if the average deviation during 24 hours is 3.8%, then these predictions perform WAY better than just guessing the same price.

Second, it sounds like you are comparing predictions to the price at that moment. The prices predicted are not the price at that specific hour - they are the average price over the course of that hour. I do state this a few times on the website, including in the main description of the charts on the home page.

Third of all, the average errors actually are an under-representation of its accuracy because it updates every hour. When it predicts an hourly price incorrectly, it readjusts predictions so that they will be more accurate.

And lastly, if I could tell you the future error of my predictions, they wouldn't really be predictions would they? They would be science fiction lol...

[Beans]

Great Thread. Nothing is Random, randomization is of course impossible as every action has a predetermined consequence. Whether the human mind or tools which the human mind has created have the processing ability to discover the results of all actions before they occur is of course unlikely at our current technological stage.

But having said that the Financial Markets are not as complex as say human life, therefor we are likely to be able to determine fairly accurate predictions of financial markets with the right input data, what of course makes these predictions less stable are variables which we do not have access to, such as other peoples influence. But again this is not random, we just don't have all the data we need.

It is by no means certain that actions have predetermined consequences, in fact it seems it's just not the case. If your right though, we had better stop wasting our time trying to build quantum computers.

Find me a theory that explains how a random activity works. For something to be random it has to happen for no reason, with nothing preceeding it. We cannot as humans beings necessarily predict the consequences of actions, but how can something happen without cause or reason?

Did I randomly come across this post and decide to start writing a response.... well if I was telling someone else about it I would probably say I came across the post randomly - however I didn't, all the events from my birth onwards led me to that point (and of course my mother and fathers actions would of also had an affect, and theirs and so on). So it wasn't random, several trillion variables which all had a set value due to other trillions of variables and so on created the unique event of me replying to this post.

But hey, its just theory of mine - I may be wrong .

:p

A universe without god needs randomness. Finding what causes randomness could be impossible, but that doesn't make it any less true. Things like radioactive decay appear to happen for no reason. Things appear random on a quantum level, while on a larger scale they don't appear random at all. It may be just a truth we have to accept unless something else comes along.

[notbatman]

Random numbers; what kind of ill concept is this?

[HCLivess]

Great Thread. Nothing is Random, randomization is of course impossible as every action has a predetermined consequence. Whether the human mind or tools which the human mind has created have the processing ability to discover the results of all actions before they occur is of course unlikely at our current technological stage.

But having said that the Financial Markets are not as complex as say human life, therefor we are likely to be able to determine fairly accurate predictions of financial markets with the right input data, what of course makes these predictions less stable are variables which we do not have access to, such as other peoples influence. But again this is not random, we just don't have all the data we need.

It is by no means certain that actions have predetermined consequences, in fact it seems it's just not the case. If your right though, we had better stop wasting our time trying to build quantum computers.

Find me a theory that explains how a random activity works. For something to be random it has to happen for no reason, with nothing preceeding it. We cannot as humans beings necessarily predict the consequences of actions, but how can something happen without cause or reason?

Did I randomly come across this post and decide to start writing a response.... well if I was telling someone else about it I would probably say I came across the post randomly - however I didn't, all the events from my birth onwards led me to that point (and of course my mother and fathers actions would of also had an affect, and theirs and so on). So it wasn't random, several trillion variables which all had a set value due to other trillions of variables and so on created the unique event of me replying to this post.

But hey, its just theory of mine - I may be wrong .

:p

A universe without god needs randomness. Finding what causes randomness could be impossible, but that doesn't make it any less true. Things like radioactive decay appear to happen for no reason. Things appear random on a quantum level, while on a larger scale they don't appear random at all. It may be just a truth we have to accept unless something else comes along.

Thank you for introducing basics to the poor-minded.

[K128kevin]

A universe without god needs randomness. Finding what causes randomness could be impossible, but that doesn't make it any less true. Things like radioactive decay appear to happen for no reason. Things appear random on a quantum level, while on a larger scale they don't appear random at all. It may be just a truth we have to accept unless something else comes along.

If anything caused randomness then it wouldn't be random. I think you are mistaking randomness with phenomena that we can't explain. Just because we don't know what causes something or because something appears random to us doesn't mean that it actually is random. In fact, human history seems to continually prove that things we attribute to magic or just randomness actually have scientific causes and are far from random.

[porcupine87]

I don't know if this method adds any value. You read about an error of 1.3% on a daily basis. This is a flaw because this only counts for ex ante predictions. Not for acutal predictions into the future. For example he made the prediction of 480$. The real price was around 440$. This is an error of 8.3%.

I watched it for a while and checked:
30.march 3pm it predicted 489$. Real price (24h later) was: 455$ (7%error)
31.3. 12am he predicted 465$. Real price was: 486$ (4.3%)
1.4. 1am he predicted 456$. Real price was: 485$ (6.3%)
1.4. 8pm he predicted 466$. Real price was: 434$ (6.9%)
2.4. 11am he predicted 449$. Real price was: 422$ (6%)

So there is an average error of about 6%.

So is that good or bad? We can just estimate that quickly. Take the prediction: "Price today is the price in 24h". This is just the volatility. I took the hourly prices of bitstamp and compared it to the value 24h later. The mean error was 3.8%.

-> So if you just guess that the current price is the price in 24h you drive better than with this prediction of the neural network (although I just made a sample of 6 predictions of it).

There are several things wrong with this data you are providing.

First of all, I'm not sure how you chose the specific times you tested, but they are clearly it's worst predictions. As you said, you only took a sample of 6 predictions. Every hour I take a sample of something like 60,000 predictions and calculate the average error. It comes to about 1.3%, so actually that is the correct figure (not 6%). Furthermore, this means that if the average deviation during 24 hours is 3.8%, then these predictions perform WAY better than just guessing the same price.

So you saying you run your neural network on learning data and then test against test data and so you come to the error of 1.3%? Then my used sample of 6 would be a pretty amazing coincidance. Because I just picked 6 random times when I just thought about it. But because not of the real future predictions is even close to 1.3%, my guess is you just use one set of data and the 1.3% is for the learning data itself.

So am I right?

Second, it sounds like you are comparing predictions to the price at that moment. The prices predicted are not the price at that specific hour - they are the average price over the course of that hour. I do state this a few times on the website, including in the main description of the charts on the home page.

Ok, I took the opening price at Bitstamp. Do you use Bitstamp prices? Anyhow, if I use the average price it will still not even get close to that 1.3%.

So, we can make the test. You use bitstamp prices and when you make a prediction at 4pm for 4pm the other day, what average price do you predict? From 3-4pm or from 4-5pm? Than we can check together? (do you save your predictions?)

[paulsonnumismatics]

Ok, 7 pages of the thread now.

The guy missed for a Dodger's Stadium wide, can we let this thread die?

Damn, noone seemed to say this.

There is no system. Maybe someday Psychohistory will be developed. Not this year. And definitely not this guy.

*Goes back to play satoshidice*

[K128kevin]

Ok, 7 pages of the thread now.

The guy missed for a Dodger's Stadium wide, can we let this thread die?

Damn, noone seemed to say this.

There is no system. Maybe someday Psychohistory will be developed. Not this year. And definitely not this guy.

*Goes back to play satoshidice*

lol really? Does it really bother you that much that this thread exists? It's not even talking about the prediction anymore (which was definitely at least partly successful in that it predicted for the price to spike upward, only it didn't spike upward quite as much as initially predicted).

And why is it that you say "this guy" definitely will not be the one to develop a successful system for predicting bitcoin prices? It seems like "this guy" already has.

[bryant.coleman]

Extremely confusing. Here is why:

a) 24-hour prediction for 10 pm 11th April 2014 to 10 pm 12th Apr : Bitcoin will rise from $422 to $470+, and then go back to sub-400 levels.
b) 5-day prediction, separated for the same time period as above: Bitcoin fairly stable, at $403 to $406. No spikes, no dips.

How can the prediction differ, for the same time-period?

[K128kevin]

So you saying you run your neural network on learning data and then test against test data and so you come to the error of 1.3%? Then my used sample of 6 would be a pretty amazing coincidance. Because I just picked 6 random times when I just thought about it. But because not of the real future predictions is even close to 1.3%, my guess is you just use one set of data and the 1.3% is for the learning data itself.

So am I right?

Ok, I took the opening price at Bitstamp. Do you use Bitstamp prices? Anyhow, if I use the average price it will still not even get close to that 1.3%.

So, we can make the test. You use bitstamp prices and when you make a prediction at 4pm for 4pm the other day, what average price do you predict? From 3-4pm or from 4-5pm? Than we can check together? (do you save your predictions?)

1. Yes you are right. I use the same ~4 million transactions to train as to test. Training on the entire history of transactions makes the neural network more accurate, and testing on the same data it is trained on does not reduce accuracy at all because the data is so large that it would be impossible for the neural network to attain an average error of 1.3% without having the ability to generalize.

The problem of testing on a training set is only relevant when the neural network is specifying to the training data and therefore, incapable of generalizing. I don't know how much you know about neural networks, but a 3-layer neural network can represent any continuous function, but nothing beyond that. The bitcoin historic data has so many transactions that no continuous function could possibly be specified to the data. In order to attain any reasonable error it would have to generalize.

2. It sounds like you didn't choose times randomly at all - you chose times when you were thinking about it, which were probably times when news was affecting the price or just times when the predictions were actually wrong. And if it were truly random, then yes, it would be a coincidence (not coincidance).

3. You would know that I use Bitstamp prices if you had read the first line of text on my website (besides the follow on twitter/like on facebook bit). But anyway, you shouldn't be trying to run these "studies" on my data if you haven't even read about what anything on the website is.

4. Also this is probably pretty self-evident but taking 6 data points to try to determine accuracy is like rolling a 6 sided die 2 times to see what percent of the time it lands on each side. It's not even remotely close to accurate and you shouldn't frame it to be so.

*Sorry if I'm being kind of a dick here, but it just annoys me when people take 6 bad predictions and then claim that this is indicative of the entire system's accuracy. Especially when I am currently presenting this data for free, and it is likely much more accurate than any human would be able to predict.

Also yes I do save my predictions, and you will be able to see charts with predictions matched up against actual data hopefully in the next few days. I have a lot of work coming up though so we'll see.

[K128kevin]

Extremely confusing. Here is why:

a) 24-hour prediction for 10 pm 11th April 2014 to 10 pm 12th Apr : Bitcoin will rise from $422 to $470+, and then go back to sub-400 levels.
b) 5-day prediction, separated for the same time period as above: Bitcoin fairly stable, at $403 to $406. No spikes, no dips.

How can the prediction differ, for the same time-period?

The reason they differ right now is that the 5-day prediction looks at averages over 6 hour periods. Because of this, it doesn't really know about the returning rise in price that took place recently, so it hasn't been able to update accordingly. I'm currently working on having the 5-day prediction look at smaller time intervals but it is difficult because this greatly increases run time.

Basically they are just two different neural networks with 2 different sets of inputs. Most of the time they pretty much agree, but sometimes there are disparities between them (like now).

[TwinWinNerD]

In reality OPs site is just using a brownsh random walk generator with a monthly expected return of 2-10 %



just kiddin

[bryant.coleman]

The reason they differ right now is that the 5-day prediction looks at averages over 6 hour periods. Because of this, it doesn't really know about the returning rise in price that took place recently, so it hasn't been able to update accordingly. I'm currently working on having the 5-day prediction look at smaller time intervals but it is difficult because this greatly increases run time.

OK... thanks for the explanation. BTC is rising now, it has reached $425. Let's see whether it will reach $475, as you had predicted.

[LordMo]

Ok, 7 pages of the thread now.

The guy missed for a Dodger's Stadium wide, can we let this thread die?

Damn, noone seemed to say this.

There is no system. Maybe someday Psychohistory will be developed. Not this year. And definitely not this guy.

*Goes back to play satoshidice*

Sounds like a fun and educational project for a student. Hopefully, that's all it was.

[K128kevin]

OK... thanks for the explanation. BTC is rising now, it has reached $425. Let's see whether it will reach $475, as you had predicted.

I think that the neural network often overestimates the magnitude of spikes like this. My guess would be that it will reach about $460 but we'll see.

[porcupine87]

So you saying you run your neural network on learning data and then test against test data and so you come to the error of 1.3%? Then my used sample of 6 would be a pretty amazing coincidance. Because I just picked 6 random times when I just thought about it. But because not of the real future predictions is even close to 1.3%, my guess is you just use one set of data and the 1.3% is for the learning data itself.

So am I right?

Ok, I took the opening price at Bitstamp. Do you use Bitstamp prices? Anyhow, if I use the average price it will still not even get close to that 1.3%.

So, we can make the test. You use bitstamp prices and when you make a prediction at 4pm for 4pm the other day, what average price do you predict? From 3-4pm or from 4-5pm? Than we can check together? (do you save your predictions?)

1. Yes you are right. I use the same ~4 million transactions to train as to test. Training on the entire history of transactions makes the neural network more accurate, and testing on the same data it is trained on does not reduce accuracy at all because the data is so large that it would be impossible for the neural network to attain an average error of 1.3% without having the ability to generalize.

The problem of testing on a training set is only relevant when the neural network is specifying to the training data and therefore, incapable of generalizing. I don't know how much you know about neural networks, but a 3-layer neural network can represent any continuous function, but nothing beyond that. The bitcoin historic data has so many transactions that no continuous function could possibly be specified to the data. In order to attain any reasonable error it would have to generalize.

So did you test your prediction against real future prices? Because that is what counts. To make a 3 layer neural network is not rocket science. But I don't know what input data you use. Only the 4mio. transactions? (where did you get them?)

*Sorry if I'm being kind of a dick here, but it just annoys me when people take 6 bad predictions and then claim that this is indicative of the entire system's accuracy. Especially when I am currently presenting this data for free, and it is likely much more accurate than any human would be able to predict.

I just used random times. Not news based. You saw the times I posted. I don't think they were news based(?) (+ the prediction you made in this thread). I used to work for a company to make predictions about sales and now I work a master thesis about nearly the same. I think I have some idea about statistics. You know about the normal distribution? 6 predictions 500% over the average estimate. Really? This is nearly unlikely like a lottery ticket.

Ok, another one. 1pm. 1pm it estimates 380$ for tomorrow. The 1.3% range is 375 and 385$. Let's see!

PS: You should use test data. Take your network for 1.4 backwards and test it to predict the price of 2.4. and so on. Then you get real prediction (maybe you should not use the last 2 voltatile days)

[LordMo]

Ok, another one. 1pm. 1pm it estimates 380$ for tomorrow. The 1.3% range is 375 and 385$. Let's see!

The Lord predicts it'll stay above $400 and that the above prediction is wrong.

[Beans]

A universe without god needs randomness. Finding what causes randomness could be impossible, but that doesn't make it any less true. Things like radioactive decay appear to happen for no reason. Things appear random on a quantum level, while on a larger scale they don't appear random at all. It may be just a truth we have to accept unless something else comes along.

If anything caused randomness then it wouldn't be random. I think you are mistaking randomness with phenomena that we can't explain. Just because we don't know what causes something or because something appears random to us doesn't mean that it actually is random. In fact, human history seems to continually prove that things we attribute to magic or just randomness actually have scientific causes and are far from random.

You could be correct. I'm not confusing anything, I'm just reluctant to say it's impossible to know why it happens. Even though it probably is. I wouldn't blindly accept a non random universe while all the evidence points in the other directions either. Even if the universe started from a single particle. Where did that particle come from? There's only two answers. I'm pretty sure I know which one you picked.

[K128kevin]

So did you test your prediction against real future prices? Because that is what counts. To make a 3 layer neural network is not rocket science. But I don't know what input data you use. Only the 4mio. transactions? (where did you get them?)

I just used random times. Not news based. You saw the times I posted. I don't think they were news based(?) (+ the prediction you made in this thread). I used to work for a company to make predictions about sales and now I work a master thesis about nearly the same. I think I have some idea about statistics. You know about the normal distribution? 6 predictions 500% over the average estimate. Really? This is nearly unlikely like a lottery ticket.

Ok, another one. 1pm. 1pm it estimates 380$ for tomorrow. The 1.3% range is 375 and 385$. Let's see!

PS: You should use test data. Take your network for 1.4 backwards and test it to predict the price of 2.4. and so on. Then you get real prediction (maybe you should not use the last 2 voltatile days)

Once again, you are asking questions that you would know the answer to if you read my site. You can see where I originally got the data from on the website (and I've been updating it using the bitstamp API since then).

Also making a 3-layer neural network is not rocket science but it's definitely very far from a simple task as well. Even a 2 layer neural network is complicated. Any number of layers above that, though, would be of equal difficulty to create.

Pretty soon I'm going to have charts up showing actual prices vs predicted prices. If you really want to do some measurement yourself though, you obviously can't just pick random times. You have to look at all 24 predictions from multiple 24 hour predictions and compare them to the actual prices, and take the average of those errors. In order to really get a solid measurement of this system's accuracy you would have to do this (as in test 24 predictions) at least like 40 or 50 times (and this is still a very small sample) and it would have to be over the course of at least a few weeks. Honestly, this would give you a tenuous grasp of the system's accuracy at best. 6 "random" predictions gives you absolutely zero indication of its accuracy.

[Costanza1]

But you see how unlikely it is that 6 random predictions are off by 500% of the average error right?

[K128kevin2]

But you see how unlikely it is that 6 random predictions are off by 500% of the average error right?

No because 1, it's only 6 predictions, 2, they were in pretty close proximity and it doesn't seem like they were "random". It's really not improbable at all.

[Bit_Happy]

Ok, another one. 1pm. 1pm it estimates 380$ for tomorrow. The 1.3% range is 375 and 385$. Let's see!

The Lord predicts it'll stay above $400 and that the above prediction is wrong.

The trend has turned up, so staying above $400 is highly likely. 

[porcupine87]

So did you test your prediction against real future prices? Because that is what counts. To make a 3 layer neural network is not rocket science. But I don't know what input data you use. Only the 4mio. transactions? (where did you get them?)

I just used random times. Not news based. You saw the times I posted. I don't think they were news based(?) (+ the prediction you made in this thread). I used to work for a company to make predictions about sales and now I work a master thesis about nearly the same. I think I have some idea about statistics. You know about the normal distribution? 6 predictions 500% over the average estimate. Really? This is nearly unlikely like a lottery ticket.

Ok, another one. 1pm. 1pm it estimates 380$ for tomorrow. The 1.3% range is 375 and 385$. Let's see!

PS: You should use test data. Take your network for 1.4 backwards and test it to predict the price of 2.4. and so on. Then you get real prediction (maybe you should not use the last 2 voltatile days)

Once again, you are asking questions that you would know the answer to if you read my site. You can see where I originally got the data from on the website (and I've been updating it using the bitstamp API since then).

Also making a 3-layer neural network is not rocket science but it's definitely very far from a simple task as well. Even a 2 layer neural network is complicated. Any number of layers above that, though, would be of equal difficulty to create.

Pretty soon I'm going to have charts up showing actual prices vs predicted prices. If you really want to do some measurement yourself though, you obviously can't just pick random times. You have to look at all 24 predictions from multiple 24 hour predictions and compare them to the actual prices, and take the average of those errors. In order to really get a solid measurement of this system's accuracy you would have to do this (as in test 24 predictions) at least like 40 or 50 times (and this is still a very small sample) and it would have to be over the course of at least a few weeks. Honestly, this would give you a tenuous grasp of the system's accuracy at best. 6 "random" predictions gives you absolutely zero indication of its accuracy.

Ok, random #7:
Your said, we get to 375 to 385. Real price was 425$. Error: 10.5%
Counter prediction: "price now will be price in 24hours": Error: 1.2%

If you know statitics and assume a normal distribution, you know how unlikely your real prediction error is only 1.3%. It is rather like 5-7% and worse like the counter prediction. 7 random tests are a lot. 7 times an 500% error in a row.
Again: I don't say 7 random predictions are enough to to measure your accurancy. I just say disprove the average error of 1.3% for real predictions.

Ok, another one:
Current price: 414.3$
Your neural network predicts: 417$

The good thing is, that this time, the counter prediction(same price in 24h) will be not really better than your neural network prediction

[cech4204a]

I'm not sure how you did it, but i guess you were wrong with estimation.

[AT101ET]

It seems like your system is defective. Do you have any actual algorithms or sequential mathematical programmes that are running the system or is this just a coin toss decision that you make 

[K128kevin2]

Ok, random #7:
Your said, we get to 375 to 385. Real price was 425$. Error: 10.5%
Counter prediction: "price now will be price in 24hours": Error: 1.2%

If you know statitics and assume a normal distribution, you know how unlikely your real prediction error is only 1.3%. It is rather like 5-7% and worse like the counter prediction. 7 random tests are a lot. 7 times an 500% error in a row.
Again: I don't say 7 random predictions are enough to to measure your accurancy. I just say disprove the average error of 1.3% for real predictions.

Ok, another one:
Current price: 414.3$
Your neural network predicts: 417$

The good thing is, that this time, the counter prediction(same price in 24h) will be not really better than your neural network prediction

You're obviously not reading anything that I'm writing at all. Here is an equally biased method of measuring error to yours: at 11am EST, it says the price will be $410.1. Let's see how far off that is. Then I'll measure the prediction for one hour after that and we'll see how far off it is. I'll average those two numbers and say that is the average error of the entire system.

Read what I wrote in the previous post. Your "data" means nothing, and the fact that you refuse to read my posts makes me feel so much less inclined to try to help you understand what you are doing wrong. I just don't think I should waste my time trying to explain these things to people who are unwilling to listen.

And once again, I don't see how you can take 6 measurements, compare them to my 40,000 or so (recalculated every hour, so really I've taken somewhere around 30 million measurements), and say that yours are more accurate than mine. The ONLY way you can do any reasonable test of accuracy would be if you did what I describe in the previous post. Until you do, please don't post here with your "random" measurements claiming that my neural network (which will almost definitely predict more accurately than any human) has an average error of 5+%. Your methods of calculations are terrible and your data is misleading and just straight up false. I don't believe that you know anything about statistics.

It seems like your system is defective. Do you have any actual algorithms or sequential mathematical programmes that are running the system or is this just a coin toss decision that you make  Huh

You are on the same boat porcupine87 and I would highly recommend getting off. Read the website before coming to these wild conclusions, and actually watch predictions for a while before concluding about accuracy. It was pretty accurate yesterday, although not the most accurate it has ever been. You can find out how it works by reading earlier in this post, on the website, or any number of other places.

[K128kevin2]

Hmm so the price it predicted for 11am on 4/13 was 410.1 and the actual price was 412.5. That's an error of 0.5%. Let's see what the error is at 12pm EST and average those two numbers, which, of course, will be indicative of the entire system's accuracy.

EDIT: It predicted 410.8, actual was 411.8. That's a 0.2% error.

So in conclusion, my study has definitively PROVEN that the average error of ALL predictions made by this system is 0.35%

(and for those of you who haven't realized this, I'm not serious - I'm proving a point)

OH and also the following hour it predicted 405.5 and the price was 405.6, so that's a 0.02% error, making the new average 0.23%

[jamesc760]

A complete failure from yesterday's predictions, kevin. All day long yesterday, your neural network predicted a 24-hour downward trend to $400. All day long yesterday, price stayed at or above $450. I sold 0.5 btc at 3PM thinking it would go lower. Oopsie. Not a big deal and you have stated NOT to rely on the neural network for REAL trades, so it's all my bad. Let this be a lesson, don't make trades based solely on the prediction. It's quite often wrong.

[TwinWinNerD]

There was 2 days in the last 10 days that were even close to be nearly the same. Hmm 1.3% error...

[Bit_Happy]

Finally made it over $500.
Timing is everything. 

[K128kevin2]

A complete failure from yesterday's predictions, kevin. All day long yesterday, your neural network predicted a 24-hour downward trend to $400. All day long yesterday, price stayed at or above $450. I sold 0.5 btc at 3PM thinking it would go lower. Oopsie. Not a big deal and you have stated NOT to rely on the neural network for REAL trades, so it's all my bad. Let this be a lesson, don't make trades based solely on the prediction. It's quite often wrong.

Sorry :/ But yeah that was probably it's worst prediction it has had so far in the ~1 month that it has been up. Hopefully these events will be rare.

Finally made it over $500.
Timing is everything.  Wink

Yes it did! This prediction was a wild success and occurred exactly as I predicted xD

[Bit_Happy]

Pretty quiet right now.
I think the next move is $550 and towards $575.

[atp1916]

I say....as the near future arrives day by day the price will creep back up to somewhere between 775-825...and people will wake up that day and be like DANG!

 

[K128kevin2]

I say....as the near future arrives day by day the price will creep back up to somewhere between 775-825...and people will wake up that day and be like DANG!

 

I definitely hope you are right! I don't think it will slowly creep up though. By the time it hits 700-800, I think it will be during a large spike that will take us over 1k. Idk when this will happen but I think given bitcoin's history, it seems inevitable.

[porcupine87]

Ok, random #7:
Your said, we get to 375 to 385. Real price was 425$. Error: 10.5%
Counter prediction: "price now will be price in 24hours": Error: 1.2%

If you know statitics and assume a normal distribution, you know how unlikely your real prediction error is only 1.3%. It is rather like 5-7% and worse like the counter prediction. 7 random tests are a lot. 7 times an 500% error in a row.
Again: I don't say 7 random predictions are enough to to measure your accurancy. I just say disprove the average error of 1.3% for real predictions.

Ok, another one:
Current price: 414.3$
Your neural network predicts: 417$

The good thing is, that this time, the counter prediction(same price in 24h) will be not really better than your neural network prediction

You're obviously not reading anything that I'm writing at all. Here is an equally biased method of measuring error to yours: at 11am EST, it says the price will be $410.1. Let's see how far off that is. Then I'll measure the prediction for one hour after that and we'll see how far off it is. I'll average those two numbers and say that is the average error of the entire system.

Please, can we stay at UTC. I am not from the east cost. In Germany we have summer time, so 2 hours instead of 1 hour off. This is confusing enough.

So your system predicted 417$ for 1pm 14.April. Let's look a the actual price: Oh, over $450. So the error was 8%.
In this times, when the price is not moving, you error might be really low. But same with the prediction "price in 24h is the same like current price".

Look at the 24h predictions of the last 4 days (so 72 predictions). The average error was 1.05%.

New prediction:
23.4.2014 6pm. Current price 485$. Your prediction for 24.4.2014 is 494$.

And once again, I don't see how you can take 6 measurements, compare them to my 40,000 or so (recalculated every hour, so really I've taken somewhere around 30 million measurements), and say that yours are more accurate than mine. The ONLY way you can do any reasonable test of accuracy would be if you did what I describe in the previous post. Until you do, please don't post here with your "random" measurements claiming that my neural network (which will almost definitely predict more accurately than any human) has an average error of 5+%. Your methods of calculations are terrible and your data is misleading and just straight up false. I don't believe that you know anything about statistics.

No it isn't. What is your standard deviation or the measurement error? I don't have that data. I am not on your page every hour. But fair enough. With stable prices your predictions seems to result in low errors. I know I cannot proof that I took 7 random samples. The last 8% error was in fact random like you could see. But I know it was random. And 6 measurements 500% over the claimed average error is huge and make it highly unlikely that your average error is that low.

Let's compare that. For sure there is no normal distribution because your error can only be 1.2% off in one direction and limitless in the other direction. So we can compare this more like the income distribution of US households. The median income level is about 50 000$. So every second random picked household should get us an income below 50k. Over 250 000$ accounts only for 2.3%. So let's pick 6 different random households and let's see how much they earn.
-> All households over 250 000$, 5times more than the median. How high is the possibility? 0.023^6. So to be 500% off you need in average 7 billions picks to get that result. (I know median and average are different, but to be honset, this doesn't improve the odds). So please don't tell me that 6 random picks are meaningless. And the 7th was 800% off again.

But fair enough, I have no idea over how many predictions you calculate your average error. So I think over many thousands?

Ah you still don't get the think with learning and test data. You claim you made 40,000 predictions. You would need 4.5 years for that. So you didn't! I tested the real prediction into the future. You didn't!

[maok]

Great effort from the OP and proves what you can do when you have lots of spare time, however if he'll turn pro and start trading on his predictions he'll most likely be broke before a difficulty round is completed. Even though randomness might not exists, it is near impossible to predict something you don't know and/or understand, I'm not talking about the OP skills in creating projects in his neural network but more about the unknowns of real life which will always be unpredictable and will cause price shifts which your neural network hasn't previously encountered eg. mtgox.

Also I find the 1.3% margin hilarious as even in the elections same day forecast by some serious statisticians the margin of error is around 2% and thats pretty simple stuff, while bitcoin price fluctuations are very complex, you might get it right few times but you'll get it wrong most of the times. Even if you'll collect 1 billion of trillion of data you still won't be able to accurately predict an outcome there are too many unknowns, maybe only when total bitcoin in circulation will go past 90% you'll be able to be more efficient because the market will be formed mainly by traders, which actions are somewhat predictable but at the moment you have lots of speculators and miners running the markets not only traders.

[toxic1978]

Is it me or are many out there taking this way to seriously, its a 'Predictive' informative based site. Its a experimental service which you can choose to use or not to use at your own will. The poor developer never claimed to know the future or guarantee exact predictions.

The bitcoin market is always going to be prone to unknown factors due to the media and unknown personal opinion or motive, however this will overtime become less of a problem as the market stabilizes .

It would not be impossible though to incorporate media news into the prediction Algorithm, however it may require some manual intervention for defining media results with a negative to positive scoring system. Future events and important dates can also reflect the value, but again this could be incorporated but would require human intervention to collate such events and rate them in 'affect factor'

Anyone why don't people give more advice and support to the developer rather than mock and cuss at every negative result the prediction may display.

[K128kevin2]

Is it me or are many out there taking this way to seriously, its a 'Predictive' informative based site. Its a experimental service which you can choose to use or not to use at your own will. The poor developer never claimed to know the future or guarantee exact predictions.

The bitcoin market is always going to be prone to unknown factors due to the media and unknown personal opinion or motive, however this will overtime become less of a problem as the market stabilizes .

It would not be impossible though to incorporate media news into the prediction Algorithm, however it may require some manual intervention for defining media results with a negative to positive scoring system. Future events and important dates can also reflect the value, but again this could be incorporated but would require human intervention to collate such events and rate them in 'affect factor'

Anyone why don't people give more advice and support to the developer rather than mock and cuss at every negative result the prediction may display.

Thank you! Like you said, I never claimed that it is going to guarantee accurate results or even results good enough to make a profit. It's an experiment and I'm still figuring these things out and working on improving accuracy. There are multiple warnings on the site about this as well.

And for those who are doubting about the 1.3% average error, do you really think I'm just making up numbers every hour for the average accuracy and displaying them above the charts? Those ARE the average errors whether you believe it or not. I'm not lying here.

[Cassius]

Have you considered adding in (or adding separately into another graph) other indicators - whether those are MACD or something else like Google trends figures, Twitter mentions, etc?
As I recall I think there's some kind of long-term correlation between Google trends and price.

[K128kevin2]

Have you considered adding in (or adding separately into another graph) other indicators - whether those are MACD or something else like Google trends figures, Twitter mentions, etc?
As I recall I think there's some kind of long-term correlation between Google trends and price.

I am currently planning on improving the current graphs with a more complicated system of inputs, part of which will include volume.

As far as the types of indicators that you have mentioned, I would like to eventually compile data from twitter and/or other news sources and use this data as an indicator. I believe that if done properly, this could work extremely well. However it is obviously difficult. I would also need to collect an enormous amount of data before starting. I plan on working on this in the future though for sure!

[counter]

Well I always appreciate some analysis.  If it were easy everyone would be doing it. 

[K128kevin2]

I hope the current 20 day prediction is way off!

Haha yeah I hope so too - that neural network has a pretty high error rate so there's a solid chance it's wrong. Hopefully we'll see a rise soon, but it will be interesting if it actually is right. Either way, I'm pretty confident that bitcoin will rise over $1,000 again at some point, even if it means falling down to low $300s or lower first.

[bitsmichel]

I hope the current 20 day prediction is way off!

If I understand correctly it uses historical data, but with bitcoin, there is not so much historical data.

[K128kevin2]

If I understand correctly it uses historical data, but with bitcoin, there is not so much historical data.

That's true, bitcoin is still young. There's about 3 years worth of transaction data though, something around 8 million transactions.

[cech4204a]

you are not accurate at all, like someone said, looking at todays price is not the same as looking at future, you can not predict price, when will you realise that ? this is not sports and you know your odds (proximately) and put a bet on whatever you think might win. This is financial market, where price is only about supply and demand. We can estimate supply, but you will never know the demand.

[fancyboy]

you are not accurate at all, like someone said, looking at todays price is not the same as looking at future, you can not predict price, when will you realise that ? this is not sports and you know your odds (proximately) and put a bet on whatever you think might win. This is financial market, where price is only about supply and demand. We can estimate supply, but you will never know the demand.

Agree! In case with BTC, even if you have made some analyses to predict future price is like to perform a rain dance.

[K128kevin2]

you are not accurate at all, like someone said, looking at todays price is not the same as looking at future, you can not predict price, when will you realise that ? this is not sports and you know your odds (proximately) and put a bet on whatever you think might win. This is financial market, where price is only about supply and demand. We can estimate supply, but you will never know the demand.

Historic prices are absolutely an indicator of future prices. This system is not perfect and it will certainly be wrong sometimes, but it is absolutely possible to predict prices reasonably well. Prices fluctuations aren't random - they occur in patterns. Regardless of what causes these patterns, they are recurring patterns which means they are predictable. Just because you and I can't see these patterns when we look at the data doesn't mean prices are random. And anything that is not random is predictable (i.e. everything). The neural network I have here certainly has room for improvement and I'm working on doing that now, but it still does reasonably well - better than any human would come close to doing.

[r34tr783tr78]

reality is what doesn't go away when you stop believing it
Or, it's a damn luck when the facts don't go our way.

[dunchy]

I don't think this works. But anyway, I like your attitude. Read some of the comments from your website and I see you're really open to describe the algo. Thanks!


What are  other methods that might be applicable for this ? Can you recommend some literature or place on the web ?

[K128kevin2]

I don't think this works. But anyway, I like your attitude. Read some of the comments from your website and I see you're really open to describe the algo. Thanks!


What are  other methods that might be applicable for this ? Can you recommend some literature or place on the web ?

Haha well I'm glad you are supportive!

Another method that has been thrown around, which would require a significant period of time collecting data, would be to incorporate data from news sources. I believe that keeping track of the number of certain types of words that appear on Twitter or in other places on the internet could be extremely effective in predicting price movement. This data would be used in addition to historic data as inputs to the neural network. The only problem is that for this to be effective, you would need to collect data for at least half a year or so, and even then it would still be far from an ideal amount of data.

As far as literature, I learned about neural networks in a class at school, but our textbook is a pretty good source of information about them. You can find it online somewhere or torrent it for free. This is the book: http://en.wikipedia.org/wiki/Artificial_Intelligence:_A_Modern_Approach

The book talks about a lot more than just neural networks, and it's a pretty good resource about AI.

[Cassius]

Why not start with just one data stream - perhaps Twitter? See how it correlates in the broad strokes and add to it bit by bit.
Also don't forget blockchain.info charts.

[K128kevin2]

Why not start with just one data stream - perhaps Twitter? See how it correlates in the broad strokes and add to it bit by bit.
Also don't forget blockchain.info charts.

Yeah I plan to start collecting data from Twitter at some point, just haven't gotten around to it yet. Like I've said though, it will be a while before there is enough data to make reasonable predictions.

Oh and the big logistical issue is that once I write a program for collecting Twitter data, I need something for it to run on 24/7. Might just have to buy another VPS.

[DustyRah]

You are an idiot to try and claim that you can predict BTC. That will be all.

[cech4204a]

when will people realise that predicting price is imposible? nobody will ever know how to predict price on independent market, thats the fact. If that would be posible, nobody would ever be on that market for trading. Trading is about hope, no certainty here. People are affected by emotions, they are irrational and so on...

[Cassius]

when will people realise that predicting price is imposible? nobody will ever know how to predict price on independent market, thats the fact. If that would be posible, nobody would ever be on that market for trading. Trading is about hope, no certainty here. People are affected by emotions, they are irrational and so on...

Predicting a given price at a given time accurately is impossible.
However, it is clearly possible to understand broad trends and even micro-trends. If you don't understand this then get yourself a book on elementary technical analysis.
None of this is rocket science or magic. It's simply about understanding what the trend is and how trends are changing before most other speculators pick up on it. A little bit like forecasting the weather from changes in air density, perhaps. If you know what you are looking for then you can gain a very good idea of what's going to happen in the near future. The weather is part of a complex system, too. But you wouldn't suggest that no one is capable of predicting with a reasonable degree of success whether it's going to rain tomorrow, assuming they're using the right tools?

[cech4204a]

when will people realise that predicting price is imposible? nobody will ever know how to predict price on independent market, thats the fact. If that would be posible, nobody would ever be on that market for trading. Trading is about hope, no certainty here. People are affected by emotions, they are irrational and so on...

Predicting a given price at a given time accurately is impossible.
However, it is clearly possible to understand broad trends and even micro-trends. If you don't understand this then get yourself a book on elementary technical analysis.
None of this is rocket science or magic. It's simply about understanding what the trend is and how trends are changing before most other speculators pick up on it. A little bit like forecasting the weather from changes in air density, perhaps. If you know what you are looking for then you can gain a very good idea of what's going to happen in the near future. The weather is part of a complex system, too. But you wouldn't suggest that no one is capable of predicting with a reasonable degree of success whether it's going to rain tomorrow, assuming they're using the right tools?

ok tell me than, wise man, will be the price in 1 week from now higher or lower than 450$ ? and please don't compare weather with market, it's not the same (btw, i'm master in finance and i know what you are talking about, but predicting trends for future ? not in BTC market).

[Asrael999]

when will people realise that predicting price is imposible? nobody will ever know how to predict price on independent market, thats the fact. If that would be posible, nobody would ever be on that market for trading. Trading is about hope, no certainty here. People are affected by emotions, they are irrational and so on...

Predicting a given price at a given time accurately is impossible.
However, it is clearly possible to understand broad trends and even micro-trends. If you don't understand this then get yourself a book on elementary technical analysis.
None of this is rocket science or magic. It's simply about understanding what the trend is and how trends are changing before most other speculators pick up on it. A little bit like forecasting the weather from changes in air density, perhaps. If you know what you are looking for then you can gain a very good idea of what's going to happen in the near future. The weather is part of a complex system, too. But you wouldn't suggest that no one is capable of predicting with a reasonable degree of success whether it's going to rain tomorrow, assuming they're using the right tools?

ok tell me than, wise man, will be the price in 1 week from now higher or lower than 450$ ? and please don't compare weather with market, it's not the same (btw, i'm master in finance and i know what you are talking about, but predicting trends for future ? not in BTC market).

Oooh a 50/50 coin toss - I will go lower as an emotional hedge.
I can be happy if the price has gone up as I own bitcoin and I can be happy if the price has gone down as I was right. 

[Cassius]

when will people realise that predicting price is imposible? nobody will ever know how to predict price on independent market, thats the fact. If that would be posible, nobody would ever be on that market for trading. Trading is about hope, no certainty here. People are affected by emotions, they are irrational and so on...

Predicting a given price at a given time accurately is impossible.
However, it is clearly possible to understand broad trends and even micro-trends. If you don't understand this then get yourself a book on elementary technical analysis.
None of this is rocket science or magic. It's simply about understanding what the trend is and how trends are changing before most other speculators pick up on it. A little bit like forecasting the weather from changes in air density, perhaps. If you know what you are looking for then you can gain a very good idea of what's going to happen in the near future. The weather is part of a complex system, too. But you wouldn't suggest that no one is capable of predicting with a reasonable degree of success whether it's going to rain tomorrow, assuming they're using the right tools?

ok tell me than, wise man, will be the price in 1 week from now higher or lower than 450$ ? and please don't compare weather with market, it's not the same (btw, i'm master in finance and i know what you are talking about, but predicting trends for future ? not in BTC market).

Well you just totally destroyed that straw man.
I don't know. I'm not a trader. Or a weatherman. If I guess correctly, will you admit I'm right? If I guess wrong, will you take this as confirmation that you were right? If not, this test might be a bit asymmetric...
But you've already said you know what I mean.
For what it's worth, it does look like we are coming to the end of a five-month downtrend in the next week or two, if we haven't already. So if I was a trader, I'd be seriously considering going long at this point. I certainly wouldn't rule out another dip first and would keep an option open for sub-400, though curiously that is looking less and less likely. I'm happy to concede that (if it doesn't happen) I had called that one wrong. But claiming that bitcoin is just totally unpredictable (do you mean random?) doesn't make sense to me. You only have to look at the chart on the front page of Stamp to get an overall sense of what's going on. Plus, as a master in Finance, I'm sure you've studied bubbles and the similarities between them.

[TrailingComet]

I will bet against

Think we will be $350 to $450 for the next week or two

[boumalo]

I will bet against

Think we will be $350 to $450 for the next week or two

The prediction was April 03 and he lost

[akujin]

when will people realise that predicting price is imposible? nobody will ever know how to predict price on independent market, thats the fact. If that would be posible, nobody would ever be on that market for trading. Trading is about hope, no certainty here. People are affected by emotions, they are irrational and so on...

Predicting a given price at a given time accurately is impossible.
However, it is clearly possible to understand broad trends and even micro-trends. If you don't understand this then get yourself a book on elementary technical analysis.
None of this is rocket science or magic. It's simply about understanding what the trend is and how trends are changing before most other speculators pick up on it. A little bit like forecasting the weather from changes in air density, perhaps. If you know what you are looking for then you can gain a very good idea of what's going to happen in the near future. The weather is part of a complex system, too. But you wouldn't suggest that no one is capable of predicting with a reasonable degree of success whether it's going to rain tomorrow, assuming they're using the right tools?

ok tell me than, wise man, will be the price in 1 week from now higher or lower than 450$ ? and please don't compare weather with market, it's not the same (btw, i'm master in finance and i know what you are talking about, but predicting trends for future ? not in BTC market).

idiot 

[cech4204a]

when will people realise that predicting price is imposible? nobody will ever know how to predict price on independent market, thats the fact. If that would be posible, nobody would ever be on that market for trading. Trading is about hope, no certainty here. People are affected by emotions, they are irrational and so on...

Predicting a given price at a given time accurately is impossible.
However, it is clearly possible to understand broad trends and even micro-trends. If you don't understand this then get yourself a book on elementary technical analysis.
None of this is rocket science or magic. It's simply about understanding what the trend is and how trends are changing before most other speculators pick up on it. A little bit like forecasting the weather from changes in air density, perhaps. If you know what you are looking for then you can gain a very good idea of what's going to happen in the near future. The weather is part of a complex system, too. But you wouldn't suggest that no one is capable of predicting with a reasonable degree of success whether it's going to rain tomorrow, assuming they're using the right tools?

ok tell me than, wise man, will be the price in 1 week from now higher or lower than 450$ ? and please don't compare weather with market, it's not the same (btw, i'm master in finance and i know what you are talking about, but predicting trends for future ? not in BTC market).

Well you just totally destroyed that straw man.
I don't know. I'm not a trader. Or a weatherman. If I guess correctly, will you admit I'm right? If I guess wrong, will you take this as confirmation that you were right? If not, this test might be a bit asymmetric...
But you've already said you know what I mean.
For what it's worth, it does look like we are coming to the end of a five-month downtrend in the next week or two, if we haven't already. So if I was a trader, I'd be seriously considering going long at this point. I certainly wouldn't rule out another dip first and would keep an option open for sub-400, though curiously that is looking less and less likely. I'm happy to concede that (if it doesn't happen) I had called that one wrong. But claiming that bitcoin is just totally unpredictable (do you mean random?) doesn't make sense to me. You only have to look at the chart on the front page of Stamp to get an overall sense of what's going on. Plus, as a master in Finance, I'm sure you've studied bubbles and the similarities between them.

Bubles are natural, even in nature bubles happen (example is population of mices on island) , here on the long run another will happen, but we can not tell when and how big one...we can just put tumb up or down thats all...and might be wrong

[Cassius]

when will people realise that predicting price is imposible? nobody will ever know how to predict price on independent market, thats the fact. If that would be posible, nobody would ever be on that market for trading. Trading is about hope, no certainty here. People are affected by emotions, they are irrational and so on...

Predicting a given price at a given time accurately is impossible.
However, it is clearly possible to understand broad trends and even micro-trends. If you don't understand this then get yourself a book on elementary technical analysis.
None of this is rocket science or magic. It's simply about understanding what the trend is and how trends are changing before most other speculators pick up on it. A little bit like forecasting the weather from changes in air density, perhaps. If you know what you are looking for then you can gain a very good idea of what's going to happen in the near future. The weather is part of a complex system, too. But you wouldn't suggest that no one is capable of predicting with a reasonable degree of success whether it's going to rain tomorrow, assuming they're using the right tools?

ok tell me than, wise man, will be the price in 1 week from now higher or lower than 450$ ? and please don't compare weather with market, it's not the same (btw, i'm master in finance and i know what you are talking about, but predicting trends for future ? not in BTC market).

Well you just totally destroyed that straw man.
I don't know. I'm not a trader. Or a weatherman. If I guess correctly, will you admit I'm right? If I guess wrong, will you take this as confirmation that you were right? If not, this test might be a bit asymmetric...
But you've already said you know what I mean.
For what it's worth, it does look like we are coming to the end of a five-month downtrend in the next week or two, if we haven't already. So if I was a trader, I'd be seriously considering going long at this point. I certainly wouldn't rule out another dip first and would keep an option open for sub-400, though curiously that is looking less and less likely. I'm happy to concede that (if it doesn't happen) I had called that one wrong. But claiming that bitcoin is just totally unpredictable (do you mean random?) doesn't make sense to me. You only have to look at the chart on the front page of Stamp to get an overall sense of what's going on. Plus, as a master in Finance, I'm sure you've studied bubbles and the similarities between them.

Bubles are natural, even in nature bubles happen (example is population of mices on island) , here on the long run another will happen, but we can not tell when and how big one...we can just put tumb up or down thats all...and might be wrong

That's the one point you come back on?!
Top of a bubble, the price goes down. Unsustainable growth, near vertical price rise -> pop, crash. Like has just happened. That I consider predictable. You, by all accounts, don't.
Come on, if you're going to challenge someone on something that is so obvious to the majority that it barely needs stating, then at least give me something to work with.

[akujin]

Don't waste your time with that idiot... Look at him harrassing our poor chartbuddy.. 

Explanation

well, we can see that even if you didn't post it....

 

[Cassius]

Don't waste your time with that idiot... Look at him harrassing our poor chartbuddy.. 

Explanation

well, we can see that even if you didn't post it....

 

I'll not dirty my hands with anyone who hassles chartbuddy.

[K128kevin2]

You can say what you want about the predictions that my site is making right now, but predicting prices is absolutely possible. I find it completely ridiculous that anybody would think otherwise. We've created technology that can make cars drive themselves, we have phones with high definition touch screens that we carry in our pockets, and we've LITERALLY gone to the moon, yet you think predicting the price of bitcoin is impossible?

Anything that has an underlying cause is predictable, and literally everything has an underlying cause, therefore literally everything is predictable. Maybe not everything is predictable in practice, but bitcoin certainly is. The major factors that affect the price are absolutely measurable - it's just a matter of collecting all of the data and using it properly.

Honestly, to think that it is impossible to predict bitcoin prices is naive. We have done so many things that are much more amazing than this.

[spooderman]

Don't waste your time with that idiot... Look at him harrassing our poor chartbuddy.. 

Explanation

well, we can see that even if you didn't post it....

 

I'll not dirty my hands with anyone who hassles chartbuddy.

I'm not your chartbuddy, chartguy!

[mikerbiker6]

It is impossible.
It is human nature to see patterns even when there aren't any.

For every dip your software predicts, there is a dip it didnt predict.
When people look at a chart and the predicted chart they see the things that are correctly predicted, but they overlook the things that weren't predicted, that is just how humans are. But when they notice a flaw, they say:"well yeah, it came pretty close, so let's not count that one".
That is why it is not going to work.

[mikerbiker6]

stock price is pretty similar to random walk.
Give up the hunt for unicorns.

[Cassius]

It is impossible.
It is human nature to see patterns even when there aren't any.

For every dip your software predicts, there is a dip it didnt predict.
When people look at a chart and the predicted chart they see the things that are correctly predicted, but they overlook the things that weren't predicted, that is just how humans are. But when they notice a flaw, they say:"well yeah, it came pretty close, so let's not count that one".
That is why it is not going to work.

True.
Also it is human nature to see patterns where there are patterns.
I guess the problem is telling between the two.
The other problem is that sometimes there just aren't enough faces and palms to facepalm.

[Malin Keshar]

maybe 500+ at 10th, but still more than 24h for that

[K128kevin2]

Just because my website isn't accurate all the time doesn't mean that predicting prices is impossible. This isn't even really an argument, it's objectively true that it is possible. Whether it can be done practically is another question, and it is still pretty clear that the answer is yes. In fact, it has been done on stock prices already on many occasions.

mikerbiker6 - You say that it is human nature to see patterns where there aren't any. First of all, can you prove that this is true? What are you basing this statement off of? Second of all, how is this relevant to being able to use software (which is not human) to predict bitcoin prices? Also what you are saying about people overlooking the false predictions not the case at all. In fact, people have MUCH more often done the opposite - they see one or two bad predictions, ignore the good ones, and make the generalization that the system performs poorly.

People here are claiming it is impossible to predict bitcoin prices because they themselves don't know how to do it and can't think of how/why any method would work (which is reasonable because most people have not learned about the types of algorithms and models that would be useful for doing something like this). However, it is not reasonable to make assumptions that it can't be done just because you dont' know how. It 100% is possible to accurately predict bitcoin prices, and you guys will see that this is the case eventually. I am working on improving my predictions, and I'm sure there are other people out there working on their own methods of prediction as well.

Lastly, as I've explained in other threads, to believe that bitcoin prices are unpredictable is to believe that they are random. Anything that is not random is predictable and has a cause. Anything that is random does not have a cause by definition. Something that happens for no reason is the definition of magic, so to say that bitcoin prices are unpredictable is literally the same as saying that price fluctuations are caused by magic. It's just nonsense.

[bitcoinsrus]

Just because my website isn't accurate all the time doesn't mean that predicting prices is impossible. This isn't even really an argument, it's objectively true that it is possible. Whether it can be done practically is another question, and it is still pretty clear that the answer is yes. In fact, it has been done on stock prices already on many occasions.

mikerbiker6 - You say that it is human nature to see patterns where there aren't any. First of all, can you prove that this is true? What are you basing this statement off of? Second of all, how is this relevant to being able to use software (which is not human) to predict bitcoin prices? Also what you are saying about people overlooking the false predictions not the case at all. In fact, people have MUCH more often done the opposite - they see one or two bad predictions, ignore the good ones, and make the generalization that the system performs poorly.

People here are claiming it is impossible to predict bitcoin prices because they themselves don't know how to do it and can't think of how/why any method would work (which is reasonable because most people have not learned about the types of algorithms and models that would be useful for doing something like this). However, it is not reasonable to make assumptions that it can't be done just because you dont' know how. It 100% is possible to accurately predict bitcoin prices, and you guys will see that this is the case eventually. I am working on improving my predictions, and I'm sure there are other people out there working on their own methods of prediction as well.

Lastly, as I've explained in other threads, to believe that bitcoin prices are unpredictable is to believe that they are random. Anything that is not random is predictable and has a cause. Anything that is random does not have a cause by definition. Something that happens for no reason is the definition of magic, so to say that bitcoin prices are unpredictable is literally the same as saying that price fluctuations are caused by magic. It's just nonsense.

I have followed btcpredictions for a bit and it seems on the right path.  Not saying its always correct (the weeks before leading up to last week, it originally predicted a sub 400 for today (5/8/14) (and was wondering if there would be a dip), but a few days ago the chart updated and is accurate to what it is now).

I agree if many many things are taken into account (volume, bear or bull market situation, news etc are taken into consideration: it might be possible to make an estimated guess about the price) but of course nothing (in life) is a given.  But in the long run, I like btcpredictions (and I tab it daily) and I like that it updates regularly.  Anyways, great job in creating the site and I plan to use it time and time again

[K128kevin2]

I have followed btcpredictions for a bit and it seems on the right path.  Not saying its always correct (the weeks before leading up to last week, it originally predicted a sub 400 for today (5/8/14) (and was wondering if there would be a dip), but a few days ago the chart updated and is accurate to what it is now).

I agree if many many things are taken into account (volume, bear or bull market situation, news etc are taken into consideration: it might be possible to make an estimated guess about the price) but of course nothing (in life) is a given.  But in the long run, I like btcpredictions (and I tab it daily) and I like that it updates regularly.  Anyways, great job in creating the site and I plan to use it time and time again

Hey, thanks a lot! It makes me very happy to hear that people are interested in it and following it regularly

[twiifm]

Just because my website isn't accurate all the time doesn't mean that predicting prices is impossible. This isn't even really an argument, it's objectively true that it is possible. Whether it can be done practically is another question, and it is still pretty clear that the answer is yes. In fact, it has been done on stock prices already on many occasions.

mikerbiker6 - You say that it is human nature to see patterns where there aren't any. First of all, can you prove that this is true? What are you basing this statement off of? Second of all, how is this relevant to being able to use software (which is not human) to predict bitcoin prices? Also what you are saying about people overlooking the false predictions not the case at all. In fact, people have MUCH more often done the opposite - they see one or two bad predictions, ignore the good ones, and make the generalization that the system performs poorly.

People here are claiming it is impossible to predict bitcoin prices because they themselves don't know how to do it and can't think of how/why any method would work (which is reasonable because most people have not learned about the types of algorithms and models that would be useful for doing something like this). However, it is not reasonable to make assumptions that it can't be done just because you dont' know how. It 100% is possible to accurately predict bitcoin prices, and you guys will see that this is the case eventually. I am working on improving my predictions, and I'm sure there are other people out there working on their own methods of prediction as well.

Lastly, as I've explained in other threads, to believe that bitcoin prices are unpredictable is to believe that they are random. Anything that is not random is predictable and has a cause. Anything that is random does not have a cause by definition. Something that happens for no reason is the definition of magic, so to say that bitcoin prices are unpredictable is literally the same as saying that price fluctuations are caused by magic. It's just nonsense.

You can use logic to determine markets are random walk/ unpredictable.  Markets can be broken down to individual players.  Each player only has 3 possible moves sell-buy-hold.  All players play to make profit. The players only choice is follow trend, be contrarian or neutral.  Any given day more players join or leave.  The only time anything is predictable is when you get a volatility spike or dip and volatility always revert to mean.  This usually happens on significant news.  Most experienced traders know this from trading everyday.

Or you can also just use Occam's Razor.  If markets were predictable then everyone would be doing it.  If everyone did it there would not be a market.

Since I first commented in your thread a month ago I tried to Google "neural networks" and market prediction.  Seems like this was tried for a while in late 80's wary 90s but no big firms believe in this method.  The big trading firms have all the money to hire the best scientists & mathematicians in the world.  If they don't  believe in it then it probably doesn't work.  Most of the "quants" in Wall Street are hired to calculate statistical probabilities.  They make trading decisions based on long term statistical strategies like "statistical arbitrage" or they try to determine if the derivatives are correctly priced or not using concepts from Game theory, Kelly criterion gambling, etc..  This is not the same as price prediction!  I only see some fly-by-night amateur website or youtube channel claiming to be able to predict price using "neural networks".  And from what I've seen it's just some lame technical indicator.  

What I learned about "neural networks"  is that its mainly used to teach computers to recognize patterns.  Things that humans have no problem doing like speech recognition.  Seems to me like you are misapplying "neural networks" here.  First you claim that computers are better at recognizing patterns than humans.  This first point is already debatable.  Second you argue that there are patterns in market prices.  This is the Technical Analysis argument.  There might be some patterns occurring some of the time but its not consistent enough to create a complete trading system.  And since most trading firms are moving away from technical analysis and towards quantitative analysts, it should tell you that quantitative trading is far more profitable than technical trading.  Fundamental analysis is still dominant for the "investing" world though.

[theonewhowaskazu]

You can say what you want about the predictions that my site is making right now, but predicting prices is absolutely possible. I find it completely ridiculous that anybody would think otherwise. We've created technology that can make cars drive themselves, we have phones with high definition touch screens that we carry in our pockets, and we've LITERALLY gone to the moon, yet you think predicting the price of bitcoin is impossible?

Yes because your prediction affects the outcome. Thus when testing a strategy you don't have access to the actual market conditions of when you're executing the strategy. That's why banks like derivatives so much: You can bet on a financial instrument without affecting that instrument.

[K128kevin2]

You can use logic to determine markets are random walk/ unpredictable.  Markets can be broken down to individual players.  Each player only has 3 possible moves sell-buy-hold.  All players play to make profit. The players only choice is follow trend, be contrarian or neutral.  Any given day more players join or leave.  The only time anything is predictable is when you get a volatility spike or dip and volatility always revert to mean.  This usually happens on significant news.  Most experienced traders know this from trading everyday.

Or you can also just use Occam's Razor.  If markets were predictable then everyone would be doing it.  If everyone did it there would not be a market.

Since I first commented in your thread a month ago I tried to Google "neural networks" and market prediction.  Seems like this was tried for a while in late 80's wary 90s but no big firms believe in this method.  The big trading firms have all the money to hire the best scientists & mathematicians in the world.  If they don't  believe in it then it probably doesn't work.  Most of the "quants" in Wall Street are hired to calculate statistical probabilities.  They make trading decisions based on long term statistical strategies like "statistical arbitrage" or they try to determine if the derivatives are correctly priced or not using concepts from Game theory, Kelly criterion gambling, etc..  This is not the same as price prediction!  I only see some fly-by-night amateur website or youtube channel claiming to be able to predict price using "neural networks".  And from what I've seen it's just some lame technical indicator.  

What I learned about "neural networks"  is that its mainly used to teach computers to recognize patterns.  Things that humans have no problem doing like speech recognition.  Seems to me like you are misapplying "neural networks" here.  First you claim that computers are better at recognizing patterns than humans.  This first point is already debatable.  Second you argue that there are patterns in market prices.  This is the Technical Analysis argument.  There might be some patterns occurring some of the time but its not consistent enough to create a complete trading system.  And since most trading firms are moving away from technical analysis and towards quantitative analysts, it should tell you that quantitative trading is far more profitable than technical trading.  Fundamental analysis is still dominant for the "investing" world though.

If you Googled neural networks then you should have found the numerous studies where they were used for stock price prediction successfully. Neural networks were not used much at all in the 80s. They were first created in 1943 and then barely used again until the 1990s/2000s. But this is irrelevant.

You are not the first person to make the statement "If price prediction were possible, everyone would be doing it". I'm sorry to be a dick here but this is just an extremely stupid statement. There are tons of things that are possible that not everybody is doing. One project that I was thinking of doing at one point was writing a mobile app that could recognize and identify bird calls (which there is a high demand for actually, and does not yet exist). I didn't follow through because I didn't have the data that would be required, but it is absolutely 100% doable. If we can recognize human speech and we can recognize music then we can definitely recognize bird calls. The statement that everybody would be doing this makes the false assumption that everybody knows how to do this and has the resources to do so. That is obviously false.

Furthermore, as someone who has a much better understanding of neural networks and their uses than you do, I can assure you that they do things that humans can't even come close to doing. Yes, they are used for speech recognition. They are used for computer vision as well. In fact, the first time I created a neural network was to recognize hand-written digits. It performed better than a human. Neural networks can solve the traveling sales person problem, which humans are obviously very bad at. They can be used for image compression. They can be used for countless other applications. As I've mentioned, commodity price prediction is also one of the standard uses of neural networks. This isn't something I'm inventing. However, I'm trying to be innovative in the manner in which I am applying neural networks. This is far from a "misapplication".

You argue that there are no consistent patterns in price fluctuations. The mere existence of my project 100% proves this wrong. My software finds patterns in the data, and these are patterns that no human could ever find. You are basing your arguments and assumptions here on what you see and believe as a human looking at an insurmountable amount of data. If my software was not finding consistent patterns, it would not be able to attain the average error figures that it does.

It seems like you are just dismissive of neural networks because you don't understand them. It is one thing to be a bit skeptical, but don't judge something unless you understand it. I can tell that your "research" of neural networks was extremely limited, so I recommend actually learning a bit about them if you are going to try to make an argument against their effectiveness in this application.

[twiifm]

You can use logic to determine markets are random walk/ unpredictable.  Markets can be broken down to individual players.  Each player only has 3 possible moves sell-buy-hold.  All players play to make profit. The players only choice is follow trend, be contrarian or neutral.  Any given day more players join or leave.  The only time anything is predictable is when you get a volatility spike or dip and volatility always revert to mean.  This usually happens on significant news.  Most experienced traders know this from trading everyday.

Or you can also just use Occam's Razor.  If markets were predictable then everyone would be doing it.  If everyone did it there would not be a market.

Since I first commented in your thread a month ago I tried to Google "neural networks" and market prediction.  Seems like this was tried for a while in late 80's wary 90s but no big firms believe in this method.  The big trading firms have all the money to hire the best scientists & mathematicians in the world.  If they don't  believe in it then it probably doesn't work.  Most of the "quants" in Wall Street are hired to calculate statistical probabilities.  They make trading decisions based on long term statistical strategies like "statistical arbitrage" or they try to determine if the derivatives are correctly priced or not using concepts from Game theory, Kelly criterion gambling, etc..  This is not the same as price prediction!  I only see some fly-by-night amateur website or youtube channel claiming to be able to predict price using "neural networks".  And from what I've seen it's just some lame technical indicator.  

What I learned about "neural networks"  is that its mainly used to teach computers to recognize patterns.  Things that humans have no problem doing like speech recognition.  Seems to me like you are misapplying "neural networks" here.  First you claim that computers are better at recognizing patterns than humans.  This first point is already debatable.  Second you argue that there are patterns in market prices.  This is the Technical Analysis argument.  There might be some patterns occurring some of the time but its not consistent enough to create a complete trading system.  And since most trading firms are moving away from technical analysis and towards quantitative analysts, it should tell you that quantitative trading is far more profitable than technical trading.  Fundamental analysis is still dominant for the "investing" world though.

If you Googled neural networks then you should have found the numerous studies where they were used for stock price prediction successfully. Neural networks were not used much at all in the 80s. They were first created in 1943 and then barely used again until the 1990s/2000s. But this is irrelevant.

You are not the first person to make the statement "If price prediction were possible, everyone would be doing it". I'm sorry to be a dick here but this is just an extremely stupid statement. There are tons of things that are possible that not everybody is doing. One project that I was thinking of doing at one point was writing a mobile app that could recognize and identify bird calls (which there is a high demand for actually, and does not yet exist). I didn't follow through because I didn't have the data that would be required, but it is absolutely 100% doable. If we can recognize human speech and we can recognize music then we can definitely recognize bird calls. The statement that everybody would be doing this makes the false assumption that everybody knows how to do this and has the resources to do so. That is obviously false.

Furthermore, as someone who has a much better understanding of neural networks and their uses than you do, I can assure you that they do things that humans can't even come close to doing. Yes, they are used for speech recognition. They are used for computer vision as well. In fact, the first time I created a neural network was to recognize hand-written digits. It performed better than a human. Neural networks can solve the traveling sales person problem, which humans are obviously very bad at. They can be used for image compression. They can be used for countless other applications. As I've mentioned, commodity price prediction is also one of the standard uses of neural networks. This isn't something I'm inventing. However, I'm trying to be innovative in the manner in which I am applying neural networks. This is far from a "misapplication".

You argue that there are no consistent patterns in price fluctuations. The mere existence of my project 100% proves this wrong. My software finds patterns in the data, and these are patterns that no human could ever find. You are basing your arguments and assumptions here on what you see and believe as a human looking at an insurmountable amount of data. If my software was not finding consistent patterns, it would not be able to attain the average error figures that it does.

It seems like you are just dismissive of neural networks because you don't understand them. It is one thing to be a bit skeptical, but don't judge something unless you understand it. I can tell that your "research" of neural networks was extremely limited, so I recommend actually learning a bit about them if you are going to try to make an argument against their effectiveness in this application.

Hey I'm open minded so show me a link where neural networks were used as successful trading system.

If your program can find these patterns then why are the results consistently wrong?  Either there is no pattern or the patterns are too inconsistently occurring to generate a useful prediction

[K128kevin2]

Here, I found this from literally 10 seconds of Googling, most of which was typing in the search query: http://www.cs.berkeley.edu/~akar/EE671/report_stock.pdf

The predictions are absolutely not consistently wrong, as is proven by the charts showing predicted prices against actual prices. And even if it were consistently wrong, this would be just as useful as if it were consistently right. If it were consistently wrong, how would the 24 hour prediction have an average error of under 1.3%?

Edit: Seriously if you just google this there are tons of other studies and articles on the subject as well.

[spndr7]

If your program can find these patterns then why are the results consistently wrong?  Either there is no pattern or the patterns are too inconsistently occurring to generate a useful prediction

BTC-predictions makes public, all its projected trends, in real time, that would have lead to increase the difference between projected trend and real price.

Just like, the probing electron for its position changes it velocity.

[mikerbiker6]

Untill august 2016 I am bearish on bitcoin, since its inflation is larger than the inflation of USD.
Somewhere mid 2016 the block halving happens again. Which should be positive for bitcoin.

[blatchcorn]

Untill august 2016 I am bearish on bitcoin, since its inflation is larger than the inflation of USD.
Somewhere mid 2016 the block halving happens again.

Never go full retard

[mikerbiker6]

Untill august 2016 I am bearish on bitcoin, since its inflation is larger than the inflation of USD.
Somewhere mid 2016 the block halving happens again.

Never go full retard

Please expand

[twiifm]

Here, I found this from literally 10 seconds of Googling, most of which was typing in the search query: http://www.cs.berkeley.edu/~akar/EE671/report_stock.pdf

The predictions are absolutely not consistently wrong, as is proven by the charts showing predicted prices against actual prices. And even if it were consistently wrong, this would be just as useful as if it were consistently right. If it were consistently wrong, how would the 24 hour prediction have an average error of under 1.3%?

Edit: Seriously if you just google this there are tons of other studies and articles on the subject as well.

Isn't that just a thesis paper?  The guy describes how to apply ANN to price prediction and concludes that its fairly accurate.  But nowhere in the paper he gives evidence

I looked on your site where you compare 24h results compared to actual price and none of the data points were exact.  Most of your predictions were below the actual price.  Only one or two times you got it right.  How can you claim that you predicted anything?

More importantly how can someone make money trading this info?

[K128kevin2]

Isn't that just a thesis paper?  The guy describes how to apply ANN to price prediction and concludes that its fairly accurate.  But nowhere in the paper he gives evidence

I looked on your site where you compare 24h results compared to actual price and none of the data points were exact.  Most of your predictions were below the actual price.  Only one or two times you got it right.  How can you claim that you predicted anything?

More importantly how can someone make money trading this info?

Read the paper again, they describe experiments that they ran to predict stock prices on a data set. If this study doesn't satisfy you just choose one of the dozens of others on the same subject.

You really expected the predicted prices to be EXACTLY correct? I never claimed this to be true and, in fact, the average margin of error implies that the exact opposite is the case. I've never claimed that you can make money predicting with the data on my website, nor have I claimed that it is always accurate. I've explicitly said multiple times (and on the website itself) that I don't know whether or not you could make money trading based on this model, and that the predictions are not always right.

I really don't know what point you are trying to argue here (not that this is even an argument). I've explained to you why you are wrong about prices being unpredictable and I've explained to you why we know with certainty that there are patterns in the data. It seems like you are just trying to argue about my site for the sake of arguing yet you don't really have any points, so I don't know what you are doing here.

[twiifm]

Isn't that just a thesis paper?  The guy describes how to apply ANN to price prediction and concludes that its fairly accurate.  But nowhere in the paper he gives evidence

I looked on your site where you compare 24h results compared to actual price and none of the data points were exact.  Most of your predictions were below the actual price.  Only one or two times you got it right.  How can you claim that you predicted anything?

More importantly how can someone make money trading this info?

Read the paper again, they describe experiments that they ran to predict stock prices on a data set. If this study doesn't satisfy you just choose one of the dozens of others on the same subject.

You really expected the predicted prices to be EXACTLY correct? I never claimed this to be true and, in fact, the average margin of error implies that the exact opposite is the case. I've never claimed that you can make money predicting with the data on my website, nor have I claimed that it is always accurate. I've explicitly said multiple times (and on the website itself) that I don't know whether or not you could make money trading based on this model, and that the predictions are not always right.

I really don't know what point you are trying to argue here (not that this is even an argument). I've explained to you why you are wrong about prices being unpredictable and I've explained to you why we know with certainty that there are patterns in the data. It seems like you are just trying to argue about my site for the sake of arguing yet you don't really have any points, so I don't know what you are doing here.

You haven't shown prices are predictable.  Nobody has.  If somebody could do this theyd win a Nobel Prize

[damnek]

Isn't that just a thesis paper?  The guy describes how to apply ANN to price prediction and concludes that its fairly accurate.  But nowhere in the paper he gives evidence

I looked on your site where you compare 24h results compared to actual price and none of the data points were exact.  Most of your predictions were below the actual price.  Only one or two times you got it right.  How can you claim that you predicted anything?

More importantly how can someone make money trading this info?

Read the paper again, they describe experiments that they ran to predict stock prices on a data set. If this study doesn't satisfy you just choose one of the dozens of others on the same subject.

You really expected the predicted prices to be EXACTLY correct? I never claimed this to be true and, in fact, the average margin of error implies that the exact opposite is the case. I've never claimed that you can make money predicting with the data on my website, nor have I claimed that it is always accurate. I've explicitly said multiple times (and on the website itself) that I don't know whether or not you could make money trading based on this model, and that the predictions are not always right.

I really don't know what point you are trying to argue here (not that this is even an argument). I've explained to you why you are wrong about prices being unpredictable and I've explained to you why we know with certainty that there are patterns in the data. It seems like you are just trying to argue about my site for the sake of arguing yet you don't really have any points, so I don't know what you are doing here.

You haven't shown prices are predictable.  Nobody has.  If somebody could do this theyd win a Nobel Prize

I think the belief whether markets are predictable or not can be seen as a litmus test for traders. It takes experience to understand that you cannot predict liquid markets, sorry K128kevin2. I also trained a neural network when I first started trading. Of course it didn't make me any money at all, all it did was lose because of the commissions paid..

[RandomPedestrianN9]

I think the belief whether markets are predictable or not can be seen as a litmus test for traders. It takes experience to understand that you cannot predict liquid markets, sorry K128kevin2. I also trained a neural network when I first started trading. Of course it didn't make me any money at all, all it did was lose because of the commissions paid..

Did you write one or you used someone elses? I read some research papers too and certain NN structures combined with other modules had 70%-80% accuracy, one team claimed to achieve 91% but did not disclose any details (because it was either fake or they wrote great system and kept it secret for themselves). Markets are neither random walk or logical is what i understood from materials i have read. If one could combine the right factors and systems, i believe it can at the least increase noobs trading success (say, 6-7 instead of 4-5 successful trades out of 10).

Depends what variables was your NN using, what did it count as important factor, networks structure etc...

[damnek]

I think the belief whether markets are predictable or not can be seen as a litmus test for traders. It takes experience to understand that you cannot predict liquid markets, sorry K128kevin2. I also trained a neural network when I first started trading. Of course it didn't make me any money at all, all it did was lose because of the commissions paid..

Did you write one or you used someone elses? I read some research papers too and certain NN structures combined with other modules had 70%-80% accuracy, one team claimed to achieve 91% but did not disclose any details (because it was either fake or they wrote great system and kept it secret for themselves). Markets are neither random walk or logical is what i understood from materials i have read. If one could combine the right factors and systems, i believe it can at the least increase noobs trading success (say, 6-7 instead of 4-5 successful trades out of 10).

Depends what variables was your NN using, what did it count as important factor, networks structure etc...

It was based on some python library, neural networks are pretty much a generic tool so I don't think it really matters which implementation you choose.
This guy explains the problems with overfitting really well: https://www.tastytrade.com/tt/shows/the-skinny-on-options-math/episodes/11152. Obviously neural networks are the prime candidate for overfit.

[K128kevin2]

I think the belief whether markets are predictable or not can be seen as a litmus test for traders. It takes experience to understand that you cannot predict liquid markets, sorry K128kevin2. I also trained a neural network when I first started trading. Of course it didn't make me any money at all, all it did was lose because of the commissions paid..

This is the fundamentally flawed argument that some people are making. Just because you were unable to do it doesn't mean that I can't - and it certainly doesn't mean that it is impossible. To assume that something is impossible just because you can't do it is pretty arrogant imo. Furthermore, to say that it cannot be predicted is objectively false based on the arguments that I've made earlier.

[blatchcorn]

Why is this thread still going on?

[K128kevin2]

It was based on some python library, neural networks are pretty much a generic tool so I don't think it really matters which implementation you choose.
This guy explains the problems with overfitting really well: https://www.tastytrade.com/tt/shows/the-skinny-on-options-math/episodes/11152. Obviously neural networks are the prime candidate for overfit.

Neural networks are not a generic tool. There are TONS of ways that they can be structured and trained, much of which would require designing and understanding a neural network on your own (like I have done) instead of using a library.

[damnek]

It was based on some python library, neural networks are pretty much a generic tool so I don't think it really matters which implementation you choose.
This guy explains the problems with overfitting really well: https://www.tastytrade.com/tt/shows/the-skinny-on-options-math/episodes/11152. Obviously neural networks are the prime candidate for overfit.

Neural networks are not a generic tool. There are TONS of ways that they can be structured and trained, much of which would require designing and understanding a neural network on your own (like I have done) instead of using a library.

lol you just contradicted yourself

[K128kevin2]

It was based on some python library, neural networks are pretty much a generic tool so I don't think it really matters which implementation you choose.
This guy explains the problems with overfitting really well: https://www.tastytrade.com/tt/shows/the-skinny-on-options-math/episodes/11152. Obviously neural networks are the prime candidate for overfit.

Neural networks are not a generic tool. There are TONS of ways that they can be structured and trained, much of which would require designing and understanding a neural network on your own (like I have done) instead of using a library.

lol you just contradicted yourself

um no? lol nothing here that I wrote that contradicts anything I said

[RandomPedestrianN9]

Well, feedforward and backpropagation delivered different results, then you have Kohenens maps - not many avaible materials on those. You can combine different activation functions, use different input variables (volume and closing + opening prices or volume + trend lines i donno), have another system overseeing your NN and deleting bad outputs, have another NN to predict some of your inputs....i read that S§P500 is using NN so there must be a way of making at least a little bit precise predictions with NN.

I don't think it really matters which implementation you choose.

I agree that different implementations wont probably make BIG differences but you can get 2-5% differences in errors.

[damnek]

i read that S§P500 is using NN so there must be a way of making at least a little bit precise predictions with NN.

The S&P500 is just an index that tracks the valuation of a basket of stocks.

[K128kevin2]

Well, feedforward and backpropagation delivered different results, then you have Kohenens maps - not many avaible materials on those. You can combine different activation functions, use different input variables (volume and closing + opening prices or volume + trend lines i donno), have another system overseeing your NN and deleting bad outputs, have another NN to predict some of your inputs....i read that S§P500 is using NN so there must be a way of making at least a little bit precise predictions with NN.

I don't think it really matters which implementation you choose.

I agree that different implementations wont probably make BIG differences but you can get 2-5% differences in errors.

The way that you represent your inputs can have a HUGE (way more than 2-5%) impact on accuracy. The size and number of hidden layers, of course, has a large effect too. So does the activation function. Another huge factor is the learning rate, and the strategy you use for changing learning rate as training occurs. I spent a good while on this and found that depending on what strategy I used, I would get error rates as low as 1.3% or as high as 10-20% or even higher.

[DustyRah]

um no? lol nothing here that I wrote that contradicts anything I said

You clearly said you are an idiot and that anything you say should be treated as such. Then you go on to say almost anything which seems clever so there you go, you just contradicted yourself.

Peace be with the morons~

[K128kevin2]

You clearly said you are an idiot and that anything you say should be treated as such. Then you go on to say almost anything which seems clever so there you go, you just contradicted yourself.

Peace be with the morons~

LOL

[knightcoin]

the price supposed to go down since I want to buy my btc back ....

[Malin Keshar]

more than 24 later, $20 fall, and we back to 430's. Your neural network just suffered a brain hemmorhage.


But i missed too, was expecting more than a $20 variation

[K128kevin2]

more than 24 later, $20 fall, and we back to 430's. Your neural network just suffered a brain hemmorhage.


But i missed too, was expecting more than a $20 variation

Dude this prediction was made ages ago lol... and as I recall, the result was that the price did rise when it was predicted to rise, but not quite as high as it predicted.

[DustyRah]

the price supposed to go down since I want to buy my btc back ....

Its bottom heavy, you will get it to buy back your BTC ~