AI prediction/forecasting for Bitcoin and crypto trends

[Ucy]

I will be shocked if the most sophisticated AI in the world can give you up to 20% accuracy. The only way it could succeed is if it's stealing from the one who controls the market.

[nites]

I will be shocked if the most sophisticated AI in the world can give you up to 20% accuracy. The only way it could succeed is if it's stealing from the one who controls the market.

Thats quite an interesting view you got there.
If you take a coin and flip it everytime and enter a trade, you should get ~50% correct trades, so where did you found 20%?
Could you elaborate why even most sophisticated AI would be performing worse than technical analysis or coin flip?

[nites]

Forecast for 20.07.2022 is downwards.

[tbterryboy]

If you succeed in developing such a tool, you will be rich in a short time. All you have to do is take a loan and then buy and sell, and since you will know the best time to buy and sell during the next four days, day trading for you will be very profitable.

That will actually be a brilliant idea. Taking a loan, buy and sell to make lots of profits before the loan pay back day.
But If actually this A.I has a success percentage rate of 40% to 60% then I will love to ask, please where have you been since last month from 10th to 16th June 2022 when the price of Bitcoin starting falling from $30k to $20k, because I think if this A.I tool was close to being accurate, that would have been the best time for you to have shot your shut, because am sure by now people from all around the world would have been looking for you for your A.I tool, but however, a prediction sometimes may not be accurate and that's why it's call prediction, but moreover I wish you the best as you work more on your project and come to give us update anytime soon

That is if this was true but if this was true, I don't think the op will release it in public. It's clearly said that his A.I has a 90 percent success rate, not 40 and 60 percent. Also the OP's account was registered on 14th this current month so he haven't discovered this forum yet on the dates you mentioned and maybe this invention of him is still in its development mode that time but now that it is ready, you can try it if you want to. Even the price have dip already on this current price we are at now, it's still not too late yet because the price still actively moves. You can always make a profit out of this small fluctuation we are feeling right now.

[nites]

Prediction for 21.07.2022 is downwards again.

[justdimin]

On top of prediction, almost all trading strategies and risk managment would be improved significantly.

Would you in the sea of dishonest vendors selling renamed RSI, be interested in buying real AI predictions?

What would be the guarantee that your predictions also cannot be those renamed RSI based strategies? Because, when someone arrives here, will blame all the previous ones like they are the holy person who comes here to bring all new changes but nothing changed on the profit levels so far with almost all the traders here for over the decade of time.

Honestly, I am interested on buying a trading bot or prediction or a trading strategy which will get me some 10% profits consistently but only after a year long testing. I mean even I am interested on your promotion, I will not rush to deal with you. I will take time for following you and for evaluating your strategies/bots which last for months. In the span of such evaluating time, most fake ones will disappear. This may be the reason why I am still into searching of such a profitable bot and not making profits with my sleeping bitcoins .

[nites]

On top of prediction, almost all trading strategies and risk managment would be improved significantly.

Would you in the sea of dishonest vendors selling renamed RSI, be interested in buying real AI predictions?

What would be the guarantee that your predictions also cannot be those renamed RSI based strategies? Because, when someone arrives here, will blame all the previous ones like they are the holy person who comes here to bring all new changes but nothing changed on the profit levels so far with almost all the traders here for over the decade of time.

Honestly, I am interested on buying a trading bot or prediction or a trading strategy which will get me some 10% profits consistently but only after a year long testing. I mean even I am interested on your promotion, I will not rush to deal with you. I will take time for following you and for evaluating your strategies/bots which last for months. In the span of such evaluating time, most fake ones will disappear. This may be the reason why I am still into searching of such a profitable bot and not making profits with my sleeping bitcoins .

Well, I understand your suspicious point of view, it is rightly justified. What I am doing right now is forecasting, as you probably saw, predicting next day in advance. Someone using RSI or whatever technical analysis can't do it, simply because these tools are lagging, they tell you something after the fact. If you followed forecasts for the last try days I posted, you could make 10% by simple trading in the direction of prediction, but how you trade and manage the risk is out of scope of forecasts I provide. You are welcome to follow forecasts, it started good with 3 out of 3 days direction predicted, but we will see what real accuracy is after some statistically significant number of days/forecast e.g. 30, 50 and 100 days.

[nites]

Prediction for 22.07.2022 is downwards.

[tygeade]

Okay let's test next day forecast for some time, I'll post daily predictions for the next day and make table to track the results. Will post in form of screenshot, so it is not editable after the fact.
Results will probably be worse than multi day prediction, but let's test it in real time.

If forecasting was really a way of making money in the crypto space, then we could also say that the crypto space has been converted into a gambling ground where anyone can just come in, make predictions, win a certain amount of money and leave the market.

But, whatever happens at the end of the day, I am not against you posting results tested in real time but believe me, a natural market can change predictions that may not be on target. I had multiple experiences with lots of analysts but no one stayed accurate for long period of times; not sure how one analyst needs to update themselves according to market conditions.

[stompix]

I will be shocked if the most sophisticated AI in the world can give you up to 20% accuracy. The only way it could succeed is if it's stealing from the one who controls the market.

You don't need anything sophisticated. Just this:

put in hamster in and you're going to beat S&P earnings. Proven!
Also, R.I.P. Mr Goxx, the cutest trader in the world.

Hi,

I reserach and extensively use AI and machine learning for trends prediction in bitcoin and other crypto's as well.
The model is built using cutting edge algorithms and it is capable of predicting 5 days ahead with more than 90% accuracy.

Yeah right, and why are you selling these magic golden goos rather than taking advantage of it and making yourself rich without a care in the world?
5 days ahead and 90% accuracy, common, if you would have said something at least dreamable you would have found yourself a sucker by now, throw in some genome mapping and supersymmetric theories to make it really unbelievable. 

If forecasting was really a way of making money in the crypto space, then we could also say that the crypto space has been converted into a gambling ground where anyone can just come in, make predictions, win a certain amount of money and leave the market.

If forecasting was real and with that accuracy, it will not resemble gambling at all, it will be like getting paid two-to-one odds for bets that are 1:1.1, minimizing the risks with a guaranteed 9 in 10 wins, which is not at all gambling, it's like shooting fish in a barrel and using and RPG for it. 
Crypto itself is already gambling, and probably worse than that, at least in sports you're only concerned about the adversary, there is no chance a war might reduce your funds by 75%, some twitter guy posting will make all your players miss, and it's not like a basketball match played one thousand miles away will affect the result in your boxing match.

[nites]

If forecasting was really a way of making money in the crypto space, then we could also say that the crypto space has been converted into a gambling ground where anyone can just come in, make predictions, win a certain amount of money and leave the market.

But, whatever happens at the end of the day, I am not against you posting results tested in real time but believe me, a natural market can change predictions that may not be on target. I had multiple experiences with lots of analysts but no one stayed accurate for long period of times; not sure how one analyst needs to update themselves according to market conditions.

Forecasting is a way to take gambling away from the markets. Yes, everyone can make predictions, but everyone will lose money and leave the market if they dont have tools or knowledge to forecast the market with decent accuracy.

Well, analysts are flawed beforehand, they have emotions, greed, they need to pay rent, eat, sleep, etc. That's why I built smart machine to do that. Of course market can change, hence my software is adaptive.

Yeah right, and why are you selling these magic golden goos rather than taking advantage of it and making yourself rich without a care in the world?
5 days ahead and 90% accuracy, common, if you would have said something at least dreamable you would have found yourself a sucker by now, throw in some genome mapping and supersymmetric theories to make it really unbelievable.  Wink

I'm not sure if you are familiar, but if you sell something you get money, and if you sell it a lot of times, you get rich. Crazy, huh?

Maybe it is dreamable for you to trade profitably and make a good predictions, for me it's reality. I predicted last 5 days closing candles in advance, what have you done in that time, except making 'trying to be funny' hamster trader posts.

Edit: well, look at that, today is also going to close as predicted, must be supersymmetry.

[stompix]

I'm not sure if you are familiar, but if you sell something you get money, and if you sell it a lot of times, you get rich. Crazy, huh?

Yeah just how some authors make millions selling lottery earning schemes or how some Kiyosaki made millions selling books giving financial advice when in reality he himself went bankrupt twice. Do you know what people call those?

I predicted last 5 days closing candles in advance, what have you done in that time, except making 'trying to be funny' hamster trader posts.

First, don't talk bad about Mr. Goxx, that hamster had a proven record, documented, and he has made real money with real transactions.
You said you were being able to predict 5 days in advance, your first prediction was not only off but you came also with fake numbers for the assuming real price..

Take a look, predictions I posted and real price for the 4 predicted days; 100% correlation.

19937.01                              
19381.30
20040.58
20372.07

19970.56
19323.91
20212.07
20569.92

This is the data for the days of 16-19

Jul 19, 2022   $22,467.85   $23,666.96   $21,683.41   $23,389.43   
Jul 18, 2022   $20,781.91   $22,633.03   $20,781.91   $22,485.69   
Jul 17, 2022   $21,195.04   $21,600.64   $20,778.18   $20,779.34   
Jul 16, 2022   $20,834.10   $21,514.40   $20,518.90   $21,190.32

What you claim to have happened never happened.

So since neither the opening, closing lowest, nor highest was anywhere near what you predicted.
You predicted the first day wrong, you went for lower the next one and it got up, and you missed the growth between day 2 and day 4 by a full 100%, going for 1000$ when it went up 2600$. So, it's in the range of hamster accuracy.

[nites]

Can you read properly?

This is what I wrote:

I've made a mistake with predictions I posted. I've found there is some error in the code and function for future prediction is not working properly.
Basically what happened, model predicted the test data, not the future, so technically it was not a prediction.
Currently, I'm able just to post next day forecast, until I fix for the multi days function.

It was prediction in the test, not the future dates you said. I stated that in few posts. Why are u taking it out of context?

Prediction for 22.07.2022 is downwards.

I meant on these daily predictions I post everyday.

If you are going to take everything out of context and just lie, I would like you to leave the thread.

[stompix]

Can you read properly?

Yes, I do!

This is what I wrote:

I've made a mistake with predictions I posted. I've found there is some error in the code and function for future prediction is not working properly.
Basically what happened, model predicted the test data, not the future, so technically it was not a prediction.
Currently, I'm able just to post next day forecast, until I fix for the multi days function.

Again how do you call guys that claim they have a tested system with 90% accuracy that flops from the first try?

Why are u taking it out of context?

Can't you read properly?
I already told you my problem is not with your failed prediction as I already knew those numbers will not be achievable ever, but with the fact that you quoted a correlation with the real numbers on the market which were definitely fake!
You can go back to your post where you claim that

Take a look, predictions I posted and real price for the 4 predicted days; 100% correlation.

and tell me where you got the "real prices" of 19970.56 and 19970.56 since it never went below 20k since mid-14th.

[nites]

Let me break it down for you, as I'm not yet sure if your brain is not capable of understanding simple things or you are just here to cure your frustration from not having any knowledge about the topic.

When you train and test AI, you split the data into the training and test phase in the dataset. Test phase is where you test system performance. Then you can use that to predict the future.

I posted forecast on July 15th:

Finally done,

here's the forecast for the next 4 days.

19937.01
19381.303
20040.582
20372.076

and then 3 days after I noticed I didnt made forecast but these predictions are the last 4 days in the TEST PHASE from dataset.

I've made a mistake with predictions I posted. I've found there is some error in the code and function for future prediction is not working properly.
Basically what happened, model predicted the test data, not the future, so technically it was not a prediction.
Currently, I'm able just to post next day forecast, until I fix for the multi days function.

Take a look, predictions I posted and real price for the 4 predicted days; 100% correlation.

19937.01                              
19381.30
20040.58
20372.07

19970.56
19323.91
20212.07
20569.92

Quote
Take a look, predictions I posted and real price for the 4 predicted days; 100% correlation.
and tell me where you got the "real prices" of 19970.56 and 19970.56 since it never went below 20k since mid-14th.

What you see here is prediction for the test phase, not the future dates you are so stubbornly trying to point as failed prediction. And I OPENLY said it was my mistake and it is not a FUTURE forecast.

And how many times I said in this thread I'm not trying to forecast exact prices, but the trend and direction, and somehow you failed to even read that.

After that I said also multiple times that I will forecast one day in advance until I fix multidays forecast, exactly what I am doing last 4 days. Because you can't read the thread, here's quotes of them for you.

Prediction for tomorrow 19.07.2022 is upwards.

Forecast for 20.07.2022 is downwards.

Prediction for 21.07.2022 is downwards again.

Prediction for 22.07.2022 is downwards.

Now take that, and compare it where market closed.

Once again, if you are going to notoriusly lie and take all the things out of context, please restrain yourself from commenting.
You have made cool joke, expressed your frustration and it is time to do that on the next thread, as it seems that's what are you doing in your life.

If you are somehow able to write intelligent comments like other guys here, I would be very happy to answer you.

[teosanru]

Hi,

I reserach and extensively use AI and machine learning for trends prediction in bitcoin and other crypto's as well.
The model is built using cutting edge algorithms and it is capable of predicting 5 days ahead with more than 90% accuracy.
On top of prediction, almost all trading strategies and risk managment would be improved significantly.

Would you in the sea of dishonest vendors selling renamed RSI, be interested in buying real AI predictions?
Here's photo of testing chart for the last few months:



EDIT:
One more photo where model is trained for daytrading, predicting next day in advance:

What is the basics of this? Is this just using multiple indicators and price action? Or is it factoring in all the macro economic indicators as well? I suggest you give some thesis behind this if you want someone to believe this idea. To me so far this looks imaginary let's wait for few days to see if the market still moves as per your line. But will love to know about this in detail.

[nites]

What is the basics of this? Is this just using multiple indicators and price action? Or is it factoring in all the macro economic indicators as well? I suggest you give some thesis behind this if you want someone to believe this idea. To me so far this looks imaginary let's wait for few days to see if the market still moves as per your line. But will love to know about this in detail.

It is machine learning system. Machine learning is a subset of AI. I'm not using any technical indicators or macro-economic indicators in the system.
It's using logic and algorithms from speech recognition engineering. Price is processed using various algorithms and then machine learns dependencies, patterns, basically learning how to predict it.

I will continue to post daily predictions. Let's see the result for e.g. 30 days.

Current standings:


Still can't append photo for some reason.

[nites]

Prediction for 23.07.2022 is downwards.

[nites]

Another correct prediction, currently standing at 5/5 accuracy predicting direction of daily closing candles.

I will not be able to run the system for tomorrow, so the next forecast will be for Monday, 25th.

[nites]

Prediction for 25.07.2022 is downwards.