Title: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 05, 2014, 06:27:24 PM Using data from coinmarketcap, I've created a model which can be used to predict which coins to invest in (for a daily price gain).
Thus far, the model appears to work relatively well. Yet, it has glitches. I should also point out the model isn't used to forecast which coins are "best". It's simply based on factors that lead to daily price change. I will plan to post here and see if my model is working or not. 3/5/14 Today's top 5 will be: 1. Quibit 2. Olympic coin 3. Peseta coin 4. Ultra coin 5. Riecoin Today's biggest losers will be: 1. Union coin 2. Megacoin 3. Quark 4. Protoshares 5. Astrocoin Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: PyroClasTix on March 05, 2014, 06:50:51 PM I like this. Don't edit for the sake of editing. Just keep the results no matter what. I want to see if your model works well.
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 05, 2014, 07:03:19 PM I like this. Don't edit for the sake of editing. Just keep the results no matter what. I want to see if your model works well. Cool. That's what I plan to do - no edits. I wasn't so sure I should pick just 5. I have been doing this for every coin on coinmarketcap and then ranking them. I figured I'd just choose the top 5 and bottom 5 for now. The goal will be to see if these top 5 and bottom 5 really do well/poor. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: crypto33 on March 05, 2014, 07:11:24 PM Leprocoin
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: PyroClasTix on March 05, 2014, 08:29:39 PM I like this. Don't edit for the sake of editing. Just keep the results no matter what. I want to see if your model works well. Cool. That's what I plan to do - no edits. I wasn't so sure I should pick just 5. I have been doing this for every coin on coinmarketcap and then ranking them. I figured I'd just choose the top 5 and bottom 5 for now. The goal will be to see if these top 5 and bottom 5 really do well/poor. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: qiwoman on March 05, 2014, 09:48:35 PM Can you also post what 24 hours schedule aka time zone you are doing because we are in pacific time so that is also relative to where we are as we are all all over the place. ;D ;D
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: jonnysomething on March 05, 2014, 10:10:53 PM Interesting concept. I enjoy day trading as a hobby so I'll keep an eye on this for sure.
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 06, 2014, 04:15:16 PM OK, so yesterday went pretty well. Below, I will include info on the price change from yesterday afternoon to yesterday around midnight (1st number) and price change from yesterday afternoon to today (2nd number)
PREDICTED WINNERS Riecoin 6% -3% Ultracoin 81% 258% Pesetacoin 12% -36% OlympicCoin 26% 28% QubitCoin 23% -4% AVERAGE 30% 49% PREDICTED LOSERS Unioncoin 1% 39% Megacoin 3% 0% Quark 1% 0% Protoshares 10% -1% Astrocoin 2% -4% AVERAGE 3% 7% Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: chesthing on March 06, 2014, 04:29:44 PM What info did you use to predict the ultra jump if I may ask?
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: cryptonewbie on March 06, 2014, 04:34:59 PM What info did you use to predict the ultra jump if I may ask? Yeah this is probably what everyone reading the thread is thinking. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 06, 2014, 04:49:29 PM Everything is based on a regression of historical data (over a few weeks) all obtained from coinmarketcap. So, just check out the site. It's a mix of those variables (market cap, price , supply, volume). To me, this market is "possible" to forecast because it's a currency market where supply is known. In econ, we all learn that if demand or supply increase/decrease, there will be a price change. The regression coefficients tells us how much of a change we can expect in the future across all cryptocurrencies.
If we obtained more up-to-date high-frequency data, the regression would be even more robust and I believe the predictions would be better. But, that would lead to a need for super high-frequency trading, which isn't really possible on these exchanges. So, for now, the daily data should work (on net - I still believe this is by no means full proof). For now, it's just a best guess and a writing it down here is just testing the hypothesis... nothing more. I would not gamble based on these findings until we prove it's working consistently. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 06, 2014, 04:54:42 PM Today (3/6) picks:
Winners: Tagcoin Takeicoin Grain Unioncoin Ultracoin Losers: Extremecoin Astrocoin Auroracoin Frozen Velocitycoin Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: cryptonewbie on March 06, 2014, 05:06:04 PM Today (3/6) picks: Winners: Tagcoin Takeicoin Grain Unioncoin Ultracoin Losers: Extremecoin Astrocoin Auroracoin Frozen Velocitycoin Do you have percentages also? Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: PyroClasTix on March 06, 2014, 05:06:37 PM Besides the new picks, what were the findings for yesterday?
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 06, 2014, 05:34:54 PM Besides the new picks, what were the findings for yesterday? They are posted (in one of my posts) above. Pretty decent so far. Maybe I should add that to the original post. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 06, 2014, 05:41:39 PM Today (3/6) picks: Winners: Tagcoin Takeicoin Grain Unioncoin Ultracoin Losers: Extremecoin Astrocoin Auroracoin Frozen Velocitycoin Do you have percentages also? No, I don't. I don't think it would be wise to be too exact with this because there's so much room for error. But, I can say that the models say today Ultra and Union coins have the biggest gains. My bet is that those who get the highest number from the model (ultra and union) will do best for the remainder of today, then overnight, will tank, while the others (grain, takea, tag) will pick up steam. I say this because that's what happened yesterday. The one's with the best model results (Qubit and Pesta) did well until midnight, then tanked. The others (ultra, olympic) were ok during the day, then did well overnight. This is all speculation, of course. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: PyroClasTix on March 06, 2014, 06:16:25 PM Wait, what are the percentages?
Gains/loss? Hi/lo? Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 06, 2014, 06:31:53 PM Wait, what are the percentages? Gains/loss? Hi/lo? The listed percentages (I believe you were referring to) are price changes. I was looking at the original price when I ran the model and it's change to the price at midnight of that day. The second percentage is the change from the original price when I ran the model to the price the following day (morning). Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: PyroClasTix on March 06, 2014, 06:52:41 PM Wait, what are the percentages? Gains/loss? Hi/lo? The listed percentages (I believe you were referring to) are price changes. I was looking at the original price when I ran the model and it's change to the price at midnight of that day. The second percentage is the change from the original price when I ran the model to the price the following day (morning). Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 06, 2014, 07:46:09 PM Wait, what are the percentages? Gains/loss? Hi/lo? The listed percentages (I believe you were referring to) are price changes. I was looking at the original price when I ran the model and it's change to the price at midnight of that day. The second percentage is the change from the original price when I ran the model to the price the following day (morning). No. I wish I could figure out how to post an excel table in here. I used the "Price Yesterday Afternoon" for modeling. Then I checked the price twice (midnight and this morning). The percentages refer to the change from the price yesterday afternoon to each (midnight and morning). This should explain it: Price at Yesterday Afternoon Price around Midnight Yesterday...Gain Price this Morning (ET)...Gain Riecoin $0.34000 $0.36126 6% 0.33 -3% Ultracoin $0.05300 $0.09600 81% 0.19 258% Pesetacoin $0.06600 $0.07359 12% 0.042 -36% OlympicCoin $0.00018 $0.00023 26% 0.00023 28% QubitCoin $0.00079 $0.00097 23% 0.00076 -4% Average 30% Average 49% Unioncoin $0.06600 $0.06690 1% 0.092 39% Megacoin $0.31000 $0.31778 3% 0.31 0% Quark $0.04600 $0.04634 1% 0.046 0% Protoshares $6.45000 $7.07735 10% 6.37 -1% Astrocoin $1.38000 $1.40490 2% 1.33 -4% Average 3% Average 7% Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: cryptonewbie on March 06, 2014, 08:04:07 PM What time are you doing these calculations?
When did you create the formula? Have you been investing your own money, or waiting to see the results for a few days? Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: PyroClasTix on March 06, 2014, 08:17:24 PM Wait, what are the percentages? Gains/loss? Hi/lo? The listed percentages (I believe you were referring to) are price changes. I was looking at the original price when I ran the model and it's change to the price at midnight of that day. The second percentage is the change from the original price when I ran the model to the price the following day (morning). No. I wish I could figure out how to post an excel table in here. I used the "Price Yesterday Afternoon" for modeling. Then I checked the price twice (midnight and this morning). The percentages refer to the change from the price yesterday afternoon to each (midnight and morning). This should explain it: Price at Yesterday Afternoon Price around Midnight Yesterday...Gain Price this Morning (ET)...Gain Riecoin $0.34000 $0.36126 6% 0.33 -3% Ultracoin $0.05300 $0.09600 81% 0.19 258% Pesetacoin $0.06600 $0.07359 12% 0.042 -36% OlympicCoin $0.00018 $0.00023 26% 0.00023 28% QubitCoin $0.00079 $0.00097 23% 0.00076 -4% Average 30% Average 49% Unioncoin $0.06600 $0.06690 1% 0.092 39% Megacoin $0.31000 $0.31778 3% 0.31 0% Quark $0.04600 $0.04634 1% 0.046 0% Protoshares $6.45000 $7.07735 10% 6.37 -1% Astrocoin $1.38000 $1.40490 2% 1.33 -4% Average 3% Average 7% Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 06, 2014, 09:25:41 PM What time are you doing these calculations? I'm doing them when I get into work. Unfortunately, this data needs to be pulled each day from coinmarketcap.com. So, that takes a bit, then I have to do a touch of excel work. I'd say I've been doing it around 10AM eastern. When did you create the formula? Before I created the formula, I gathered data for a week or so from coinmarketcap. Ideally, there would be a site with daily historical data for all altcoin prices that could be downloaded. That's not the case. It's a pain. Anyway, once i gathered the historical data. I ran a couple regressions on the historical data over the whole lot of coins. Once I got significance/coefficients, I moved on to forecasting using that model. I use to do this sort of thing for a living, so this is kinda second nature for me. Have you been investing your own money, or waiting to see the results for a few days? Not really. I started doing that with about $5, but decided against it. The tough part of this is you need accounts at all of the exchanges because some coins are traded here, others there, other somewhere else. Moreover, I do not trust his model myself. I certainly don't think anyone here should trust it. I would not advise anyone to use it until we know it works. So, for now, let's just keep seeing if it works. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 06, 2014, 09:31:37 PM Ok, that actually does help a lot more. Thanks! Are you going to keep doing this everyday? I'd like to know what you're using to figure this out with. I'm going to try to do it every day unless something comes up. I probably won't do it on weekends, but who knows. Like I said, I'm using the data on coinmarketcap.com. All the variables are there. I'd love to get more data, but that's all I have access to. Like for example, social media mentions. I think that could be a proxy for demand. But, I don't have that. Just coinmarketcap. The one site I'm hoping to incorporate is http://coinmarketcap.northpole.ro/ (http://coinmarketcap.northpole.ro/) ...which does what coinmarketcap does, but every 5 min. ...but, it's is a json. I'm a statistics geek, not a programmer. Prior to two weeks ago, I didn't even know what a json was. So, although I could obtain that data from the json itself and plop it into excel or my regression software, I can't store it (every five minutes) without manually doing so. I'm clueless when it comes to that sort of thing. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: turboblade on March 06, 2014, 09:37:44 PM id pay for good predictions
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: cryptonewbie on March 06, 2014, 09:56:55 PM A lot of the coins on your list have skyrocketed in anticipation of being added to Crypsty tomorrow. They bought votes, which increased customer confidence and drove price up.
I would obviously doubt you could have foreseen that was going to happen today. I would take an objective look at the final numbers tonight, and realize that may have played a large factor. I'm going to keep an eye on this thread. Could have a lot of potential. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 07, 2014, 03:04:03 PM So, the 3/6 to 3/7 predictions were not nearly as accurate as the first days predictions. If you recall back, our net percent gain on day 1 (for five winners) was +49%. Today's (for the five winners) was +1%. Not too good. Things looked better at midnight, when it was up 12%. All that said, the losers ended up up 7% (exactly like yesterday). The fact that the losers beat the winners though, means to me - this was a poor forecasting day.
NOTE - I think what's interesting is Bitcoin's performance (flat forecast day 1, down forecast day 2) may affect the underlying model. Just something to note. This might be something that should be tweaked in the model. Here's the data from Day 2: Winners: Price at Yesterday Afternoon Price around Midnight (ET time) Gain Price this Morning (ET) Gain Tagcoin $0.560000 $0.5600000 0% $0.5100000 -9% Takeicoin $0.003300 $0.0045000 36% $0.0041000 24% Grain $0.000035 $0.0000340 -3% $0.0000310 -11% UnionCoin $0.088000 $0.0880000 0% $0.0800000 -9% Ultracoin $0.200000 $0.2500000 25% $0.2200000 10% Average 12% Average 1% Losers: Extremecoin$0.1400000 $0.1500000 7% $0.180000 29% Astrocoin $1.2600000 $1.0800000 -14% $1.130000 -10% Auroracoin $18.3100000 $26.1000000 43% $20.710000 13% Frozen $0.0330000 $0.0550000 67% $0.036000 9% VelocityCoin $0.0017000 $0.0017000 0% $0.001600 -6% Average 20% Average 7% Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 07, 2014, 03:09:17 PM Today 3/7's predictions for largest daily price change:
Winners: Billioncoin Astrocoin Velocitycoin TakeiCoin Riecoin Losers: Premine FlorineCoin Catcoin Benjamins TopCoin Honorable Mentions for Winners: Thorcoin and Xivra - both are in top 10, and have negative 24 hour price gains (which to me is a good sign for growth going forward) Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: PyroClasTix on March 08, 2014, 08:11:03 AM Today 3/7's predictions for largest daily price change: How'd today go? Billion and Rie was spot on. Takei kind of, but problems with legal made it drop.Winners: Billioncoin Astrocoin Velocitycoin TakeiCoin Riecoin Losers: Premine FlorineCoin Catcoin Benjamins TopCoin Honorable Mentions for Winners: Thorcoin and Xivra - both are in top 10, and have negative 24 hour price gains (which to me is a good sign for growth going forward) Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: Adam8870 on March 08, 2014, 03:48:14 PM Thus far, the model appears to work relatively well. Yet, it has glitches. I should also point out the model isn't used to forecast which coins are "best". It's simply based on factors that lead to daily price change.
I will plan to post here and see if my model is working or not. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 08, 2014, 04:06:47 PM Today 3/7's predictions for largest daily price change: How'd today go? Billion and Rie was spot on. Takei kind of, but problems with legal made it drop.Winners: Billioncoin Astrocoin Velocitycoin TakeiCoin Riecoin Losers: Premine FlorineCoin Catcoin Benjamins TopCoin Honorable Mentions for Winners: Thorcoin and Xivra - both are in top 10, and have negative 24 hour price gains (which to me is a good sign for growth going forward) Not well. The losers actually did well relative to the winners yesterday. Although the lot was down, as was Bitcoin. Winners: Price at Yesterday Afternoon Price around Midnight (ET time) Gain Price this Morning (ET) Gain Billioncoin 0.000200000 0.00018 -10% 0.00015 -25% Astrocoin 1.130000000 1.2 6% $1.07 -5% Velocitycoin 0.001600000 0.0015 -6% $0.00 -25% TakeiCoin 0.004100000 0.00098 -76% $0.00 -85% Riecoin 0.640000000 0.5 -22% $0.42 -34% Average -22% Average -35% Losers: Premine 0.320000000 0.25 -22% $0.26 -19% FlorineCoin 0.006700000 0.0062 -7% $0.01 -4% Catcoin 0.290000000 0.27 -7% $0.28 -3% Benjamins 0.024000000 0.023 -4% 0.023 -4% TopCoin 0.000020000 0.000018 -10% 0.000019 -5% Average -10% Average -7% One interesting factor, though, is I've been comparing the "winners" daily change average for the three days I've been doing this. Right now their (the winners) daily increase is up 5% on average. I've also done this for the daily "losers" too. Their average is up (surprisingly) by 2% on average. But, not surprisingly, the winners are outpacing the losers on average. The most important stat, in my mind, however, is that bitcoin's daily average change over the same three days is down 5%. So, although there's been some hiccups, this does appear to be working. I've also been tweaking the model. Today's picks are based on the tweaked model. Winners: Digitalcoin Colossuscoin BetaCoin I0Coin Globe Losers: BatCoin Ixcoin BattleCoin Mastercoin Bytecoin Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 09, 2014, 03:24:36 PM Another good day.
Winners: Gain Digitalcoin 11% Colossuscoin 6% BetaCoin 17% I0Coin -6% Globe 12% AVERAGE 8% Losers: BatCoin -95% Ixcoin -15% BattleCoin -11% Mastercoin 6% Bytecoin -7% AVERAGE -24% Since we've begun, the winners are up 6% daily on average. The losers are down 4% daily on average. And Bitcoin is down 2% daily on average. I'm going to take today off. We'll regroup with new model winners and losers Monday AM. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: PyroClasTix on March 09, 2014, 05:57:39 PM Hey doing great so far.
Take the day off. You've done good work! :P Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 10, 2014, 04:52:37 PM Today's picks:
Winners: BatCoin Anoncoin Dimecoin Infinitecoin Colossuscoin Losers: Unobtanium Mastercoin RonPaulCoin TagCoin DuckDuckCoin Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 10, 2014, 06:38:46 PM Since posting earlier today, here's the picks status:
Winners: BatCoin -2% Anoncoin 3% Dimecoin 19% Infinitecoin 3% Colossuscoin -6% Average: 3% Losers: Unobtanium -2% Mastercoin -22% RonPaulCoin 12% TagCoin 0% DuckDuckCoin -12% Average: -5% Bitcoin: -2% Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 11, 2014, 01:43:27 PM Final status of yesterday's picks...
Winners: % Gain since Yesterday BatCoin 106% Anoncoin 19% Dimecoin -13% Infinitecoin -6% Colossuscoin -6% AVERAGE 20% Losers: % Gain since Yesterday Unobtanium -26% Mastercoin -9% RonPaulCoin -4% TagCoin 2% DuckDuckCoin 0% AVERAGE -7% Bitcoin 0% Since I've started posting picks the winners picks are up an average of 9% "daily". The losers are down an average of 7% "daily". And Bitcoin is down an average of 1% "daily". Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 11, 2014, 03:19:23 PM Today's picks:
Winners: Blakecoin Unobtanium Particle Globe Pandacoin (PND) Losers: ZenithCoin Electric Karmacoin microCoin PiggyCoin Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: todamoon on March 14, 2014, 05:25:16 PM I've been following this thread since the start with great interest.
Results would have been loads better if the "high" was recorded for each coin, but that would take more time etc which is why I guess you recorded it as you have. Please continue! :D Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 14, 2014, 06:38:26 PM I've been following this thread since the start with great interest. Results would have been loads better if the "high" was recorded for each coin, but that would take more time etc which is why I guess you recorded it as you have. Please continue! :D Yeh. I've been trying to create a site so I can post the picks up over the past few days. Truth is, for most coins (that have existed for sufficient time), an estimated daily growth rate forecast can be produced every day. But, like you said, that takes time and unfortunately I don't really have the time to do this every day then post every single coin's forecast, yet I am creating them.... for every coin and doing all of the above for free. I will keep on trying to load the model though. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 14, 2014, 07:46:26 PM To get back on the horse, here's the picks for 3/14 for the biggest price changes over the next day:
Winners: BattleCoin Premine MazaCoin Pandacoin (PND) VelocityCoin Losers: Grain BlackCoin Dimecoin ContinuumCoin ProsperCoin Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: todamoon on March 16, 2014, 10:26:49 PM Your model does seem to work.
A site sounds interesting, I will keep checking back here for updates. Keep up the good work! Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: Xardas2014 on March 16, 2014, 11:46:21 PM One thing seemingly overlooked, yet rather important: By posting your predictions on a public forum the prices can be influenced by that alone. I suggest you get some trusted community testers, keep it all private, analyze your formulas performance and adjust it, then post or sell the formula....which could be verified by the trusted community testers.
By simply posting the info as you are now, it is not really different from a P&D. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: jamieb81 on March 16, 2014, 11:59:27 PM I like your work, I'll be watching
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: gustav on March 17, 2014, 10:09:02 AM can op show us the formula?
This could also be the most sophisticated scam we will ever get to see. Imagine the guy tells you 'i am doing it based on past data' while he is actually buying and selling himself on exchange to make the thing fit. After gaining fellows like any other guru he could then command his own pumper and dumper army ... just a thought. Or why is the mathguy hanging out in this forum but is not brave enough to invest only 5$ but on the other hand has a lot of time for threads and is well educated about alts. Sorry op, didn't want to piss off, but i am overly sceptical i think - i got my reasons. Formula or it didn't happen ;) I can also give out predictions and then go on exchange and make it happen. Maybe wolong/fontas just got mathematical ;) Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: Hollowman338 on March 17, 2014, 10:30:13 AM definitely missed the boat on the blackcoin shenanigans.
I went ahead and dubbed these "Aurora-pumps" Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 17, 2014, 03:09:24 PM Gustav and Xardas,
I understand your point of view and I think it actually makes sense. In a sense, anyone who publicly picks stocks can alter preferences, and due to that, could potentially increase demand/prices of recommended stocks. The point is valid. That said, the formula should be altered repetitively (and regressions rerun repetitively) in order to adjust for new historical data. If I do post the regression results, it will be in an academic journal. But, getting back to posting results and boosting demand, I suppose that's an additional reason why this "could" work. But, what I was aiming to do was really test the theory here. Nothing more. I wasn't willing to toss too much money into this, and as the guy who created the model saying that I wouldn't bet on it (yet),... it would be crazy to invest your own money. Thus far, some days have worked, others not so much. But, on net, I think it's working to a degree, as winners are up. To give a bit of insight, I feel my formula is pretty much based on economic theory and swings. When I say swings, I mean it's not surprising that stocks (or in this case altcoins) that have huge swings would revert back to a mean. Since, it's a panel regression model, this is particularly true because we can measure "relative" swings from coin vs coin in variables. I use the variables on coinmarketcap.com to construct the model. Like I've said, I'd love to figure out how to run the model consistently and do it in an easier fashion. Meaning, I'm a statistics/econ guy, not a programmer. The regressions are easy. But, I'm manually pulling the data off coinmarketcap.com. Then, I'm adjusting the data, which is labor intensive. Then, I'm plugging in the forecasted model to create a price change for each coin. The process is a pain in the ass and it's becoming more complicated as new coins enter the market. This is one reason I haven't been able to really keep up with it as much as I would have liked. Ideally, I'd be the regression guy. That's it. That's where my experience comes from. The data pulling, the programming, etc... I'm not efficient with that. There's probably a way to do this in an easy manner if the community really wanted to get behind it. I mean, the truth is it could be automated. But, who knows if that will ever happen. Time will tell there. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 17, 2014, 05:24:31 PM Here's the model's picks for today and thier price at the time the model was ran:
Losers QQCoin 0.003 QubitCoin 0.00043 Premine 0.076 Phoenixcoin 0.0018 ProsperCoin 0.028 Winners Pxlcoin 0.014 FlappyCoin 0.0000062 Primecoin 1.22 Quark 0.03 RedCoin 0.0046 Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: todamoon on March 17, 2014, 05:58:17 PM Gustav and Xardas, I understand your point of view and I think it actually makes sense. In a sense, anyone who publicly picks stocks can alter preferences, and due to that, could potentially increase demand/prices of recommended stocks. The point is valid. That said, the formula should be altered repetitively (and regressions rerun repetitively) in order to adjust for new historical data. If I do post the regression results, it will be in an academic journal. But, getting back to posting results and boosting demand, I suppose that's an additional reason why this "could" work. But, what I was aiming to do was really test the theory here. Nothing more. I wasn't willing to toss too much money into this, and as the guy who created the model saying that I wouldn't bet on it (yet),... it would be crazy to invest your own money. Thus far, some days have worked, others not so much. But, on net, I think it's working to a degree, as winners are up. To give a bit of insight, I feel my formula is pretty much based on economic theory and swings. When I say swings, I mean it's not surprising that stocks (or in this case altcoins) that have huge swings would revert back to a mean. Since, it's a panel regression model, this is particularly true because we can measure "relative" swings from coin vs coin in variables. I use the variables on coinmarketcap.com to construct the model. Like I've said, I'd love to figure out how to run the model consistently and do it in an easier fashion. Meaning, I'm a statistics/econ guy, not a programmer. The regressions are easy. But, I'm manually pulling the data off coinmarketcap.com. Then, I'm adjusting the data, which is labor intensive. Then, I'm plugging in the forecasted model to create a price change for each coin. The process is a pain in the ass and it's becoming more complicated as new coins enter the market. This is one reason I haven't been able to really keep up with it as much as I would have liked. Ideally, I'd be the regression guy. That's it. That's where my experience comes from. The data pulling, the programming, etc... I'm not efficient with that. There's probably a way to do this in an easy manner if the community really wanted to get behind it. I mean, the truth is it could be automated. But, who knows if that will ever happen. Time will tell there. Like you I don't know much when it comes to programming etc. I am aware that you can get excel to auto pull data using macros. I don't know how to do this though, might be worth looking in to. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: gustav on March 18, 2014, 12:52:21 AM Gustav and Xardas, I understand your point of view and I think it actually makes sense. In a sense, anyone who publicly picks stocks can alter preferences, and due to that, could potentially increase demand/prices of recommended stocks. The point is valid. That said, the formula should be altered repetitively (and regressions rerun repetitively) in order to adjust for new historical data. If I do post the regression results, it will be in an academic journal. But, getting back to posting results and boosting demand, I suppose that's an additional reason why this "could" work. But, what I was aiming to do was really test the theory here. Nothing more. I wasn't willing to toss too much money into this, and as the guy who created the model saying that I wouldn't bet on it (yet),... it would be crazy to invest your own money. Thus far, some days have worked, others not so much. But, on net, I think it's working to a degree, as winners are up. To give a bit of insight, I feel my formula is pretty much based on economic theory and swings. When I say swings, I mean it's not surprising that stocks (or in this case altcoins) that have huge swings would revert back to a mean. Since, it's a panel regression model, this is particularly true because we can measure "relative" swings from coin vs coin in variables. I use the variables on coinmarketcap.com to construct the model. Like I've said, I'd love to figure out how to run the model consistently and do it in an easier fashion. Meaning, I'm a statistics/econ guy, not a programmer. The regressions are easy. But, I'm manually pulling the data off coinmarketcap.com. Then, I'm adjusting the data, which is labor intensive. Then, I'm plugging in the forecasted model to create a price change for each coin. The process is a pain in the ass and it's becoming more complicated as new coins enter the market. This is one reason I haven't been able to really keep up with it as much as I would have liked. Ideally, I'd be the regression guy. That's it. That's where my experience comes from. The data pulling, the programming, etc... I'm not efficient with that. There's probably a way to do this in an easy manner if the community really wanted to get behind it. I mean, the truth is it could be automated. But, who knows if that will ever happen. Time will tell there. please link us the articel here as soon as it is published. Since i think pricechange depends on a lot of factors of which not every one is a quantitative one. Since prices are multicausal the formula should become better the more influences are captchered in it. For example: Buy/sell depth on the books and how exactly it is distributed and changes is one thing. History data is another thing, then the overall mood of the market, then the blockrewarddecrease, then the change in networkhashrate and other thing. Then when you got it running and are giving out predictions regularly and gain attention you would have to factor in also the influence of your predictions (maybe deriven from views per day of your website) - there is many factors to consider. If you can catch the bigger picture it should be possible to get something useful from it. I will be watching this thread and waiting for the published first formula. Good luck with this. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 18, 2014, 02:03:44 AM Gustav and Xardas, I understand your point of view and I think it actually makes sense. In a sense, anyone who publicly picks stocks can alter preferences, and due to that, could potentially increase demand/prices of recommended stocks. The point is valid. That said, the formula should be altered repetitively (and regressions rerun repetitively) in order to adjust for new historical data. If I do post the regression results, it will be in an academic journal. But, getting back to posting results and boosting demand, I suppose that's an additional reason why this "could" work. But, what I was aiming to do was really test the theory here. Nothing more. I wasn't willing to toss too much money into this, and as the guy who created the model saying that I wouldn't bet on it (yet),... it would be crazy to invest your own money. Thus far, some days have worked, others not so much. But, on net, I think it's working to a degree, as winners are up. To give a bit of insight, I feel my formula is pretty much based on economic theory and swings. When I say swings, I mean it's not surprising that stocks (or in this case altcoins) that have huge swings would revert back to a mean. Since, it's a panel regression model, this is particularly true because we can measure "relative" swings from coin vs coin in variables. I use the variables on coinmarketcap.com to construct the model. Like I've said, I'd love to figure out how to run the model consistently and do it in an easier fashion. Meaning, I'm a statistics/econ guy, not a programmer. The regressions are easy. But, I'm manually pulling the data off coinmarketcap.com. Then, I'm adjusting the data, which is labor intensive. Then, I'm plugging in the forecasted model to create a price change for each coin. The process is a pain in the ass and it's becoming more complicated as new coins enter the market. This is one reason I haven't been able to really keep up with it as much as I would have liked. Ideally, I'd be the regression guy. That's it. That's where my experience comes from. The data pulling, the programming, etc... I'm not efficient with that. There's probably a way to do this in an easy manner if the community really wanted to get behind it. I mean, the truth is it could be automated. But, who knows if that will ever happen. Time will tell there. please link us the articel here as soon as it is published. Since i think pricechange depends on a lot of factors of which not every one is a quantitative one. Since prices are multicausal the formula should become better the more influences are captchered in it. For example: Buy/sell depth on the books and how exactly it is distributed and changes is one thing. History data is another thing, then the overall mood of the market, then the blockrewarddecrease, then the change in networkhashrate and other thing. Then when you got it running and are giving out predictions regularly and gain attention you would have to factor in also the influence of your predictions (maybe deriven from views per day of your website) - there is many factors to consider. If you can catch the bigger picture it should be possible to get something useful from it. I will be watching this thread and waiting for the published first formula. Good luck with this. Agree 100%. The price change probably depends on a good amount of factors. Unfortunately, all possible variables aren't available in one location where I can pull all the data for each of the 100+ coins and analyze it across all cryptocurrencies. Given time, I think this may be possible though. For now, we can try to improve the model and see what is working and what is not. Ideally, we'd get decent results with what we do have, which could only be improved upon once more data becomes more easily accessible or available. Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: gustav on March 18, 2014, 08:35:39 AM Gustav and Xardas, I understand your point of view and I think it actually makes sense. In a sense, anyone who publicly picks stocks can alter preferences, and due to that, could potentially increase demand/prices of recommended stocks. The point is valid. That said, the formula should be altered repetitively (and regressions rerun repetitively) in order to adjust for new historical data. If I do post the regression results, it will be in an academic journal. But, getting back to posting results and boosting demand, I suppose that's an additional reason why this "could" work. But, what I was aiming to do was really test the theory here. Nothing more. I wasn't willing to toss too much money into this, and as the guy who created the model saying that I wouldn't bet on it (yet),... it would be crazy to invest your own money. Thus far, some days have worked, others not so much. But, on net, I think it's working to a degree, as winners are up. To give a bit of insight, I feel my formula is pretty much based on economic theory and swings. When I say swings, I mean it's not surprising that stocks (or in this case altcoins) that have huge swings would revert back to a mean. Since, it's a panel regression model, this is particularly true because we can measure "relative" swings from coin vs coin in variables. I use the variables on coinmarketcap.com to construct the model. Like I've said, I'd love to figure out how to run the model consistently and do it in an easier fashion. Meaning, I'm a statistics/econ guy, not a programmer. The regressions are easy. But, I'm manually pulling the data off coinmarketcap.com. Then, I'm adjusting the data, which is labor intensive. Then, I'm plugging in the forecasted model to create a price change for each coin. The process is a pain in the ass and it's becoming more complicated as new coins enter the market. This is one reason I haven't been able to really keep up with it as much as I would have liked. Ideally, I'd be the regression guy. That's it. That's where my experience comes from. The data pulling, the programming, etc... I'm not efficient with that. There's probably a way to do this in an easy manner if the community really wanted to get behind it. I mean, the truth is it could be automated. But, who knows if that will ever happen. Time will tell there. please link us the articel here as soon as it is published. Since i think pricechange depends on a lot of factors of which not every one is a quantitative one. Since prices are multicausal the formula should become better the more influences are captchered in it. For example: Buy/sell depth on the books and how exactly it is distributed and changes is one thing. History data is another thing, then the overall mood of the market, then the blockrewarddecrease, then the change in networkhashrate and other thing. Then when you got it running and are giving out predictions regularly and gain attention you would have to factor in also the influence of your predictions (maybe deriven from views per day of your website) - there is many factors to consider. If you can catch the bigger picture it should be possible to get something useful from it. I will be watching this thread and waiting for the published first formula. Good luck with this. Agree 100%. The price change probably depends on a good amount of factors. Unfortunately, all possible variables aren't available in one location where I can pull all the data for each of the 100+ coins and analyze it across all cryptocurrencies. Given time, I think this may be possible though. For now, we can try to improve the model and see what is working and what is not. Ideally, we'd get decent results with what we do have, which could only be improved upon once more data becomes more easily accessible or available. i guess you could also pull your data directly from first sources instead of using coinmarketcap.com Why not use exchange-api and pull data of networks directly from the walletclients (should be possible - i don't know exactly how though. You'd need some expert on this who can tell you how to pull data from first sources). All the data can be accessed already in realtime with highest accuracy. coinmarketcap.com is a mix of data from differet exchanges. Other pages calculate slightly different numbers. I am not an expert on the technical side. You'd need possibly someone who can tell you how to pull in the most accurate data in realtime. It is possible. That will be a some effort though to set that up for each coin. I'd recommend you look for a technical expert who can tell you where to get your desired data from in the most convinient way. Data is there already - you only need to know how to access. Someone help this guy to get him accurate realtime data! Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: jamieb81 on March 18, 2014, 04:40:38 PM Keep up the good work, this is something that can get big!
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: El Dude on March 18, 2014, 04:42:45 PM winners
litecoin losers every other coin Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: barbaricmustard on March 18, 2014, 06:26:40 PM winners litecoin losers every other coin A bit closed-minded, don't you think? Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 18, 2014, 08:49:47 PM Yesterday's results were OK, nothing great:
Losers Price Yesterday Price Today Change QQCoin 0.003 0.0024 -20.00% QubitCoin 0.00043 0.00038 -11.63% Premine 0.076 0.068 -10.53% Phoenixcoin 0.0018 0.0016 -11.11% ProsperCoin 0.028 0.027 -3.57% AVERAGE -11.37% Winners Price Yesterday Price Today Change Pxlcoin 0.014 0.013 -7.14% FlappyCoin 0.0000062 0.000006 -3.23% Primecoin 1.22 1.21 -0.82% Quark 0.03 0.035 16.67% RedCoin 0.0046 0.0048 4.35% AVERAGE +1.97% Bitcoin 621.54 616.61 -.79% Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 19, 2014, 12:03:57 AM Little late today, but just ran the model here are the picks (should be interesting, the model picks Aurora as a daily loser. I know that airdrop date is upcoming, so let's see what happens with that one:
Losers Price Animecoin 0.000018 Auroracoin 18.23 Elacoin 0.18 Bitgem 2.51 BlackCoin 0.016 Winners Price Blakecoin 0.013 ContinuumCoin 0.0000032 CasinoCoin 0.0017 BatCoin 0.000006 ZenithCoin 0.0013 Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 20, 2014, 01:31:35 AM Not good results yesterday, but if it's a caveat, I noticed I made an error in data entry, which could have thrown off the model. Here's the results:
Losers Price Yesterday Price Today Change Animecoin 0.000018 0.0000061 -66.11% Auroracoin 18.23 16.07 -11.85% Elacoin 0.18 0.14 -22.22% Bitgem 2.51 2.45 -2.39% BlackCoin 0.016 0.023 43.75% AVERAGE -11.76% Winners Price Yesterday Price Today Change Blakecoin 0.013 0.013 0.00% ContinuumCoin 0.0000032 0.0000026 -18.75% CasinoCoin 0.0017 0.0018 5.88% BatCoin 0.000006 0.0000058 -3.33% ZenithCoin 0.0013 0.0013 0.00% AVERAGE -3.24% Bitcoin 612.95 607.25 -0.93% Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 20, 2014, 01:34:34 AM Here's today's picks:
Losers Price Today 66 Coin 1154 ProsperCoin 0.045 Extremecoin 0.069 BlackCoin 0.023 Pesetacoin 0.021 Winners Price Today Pandacoin (PND) 0.00000068 Premine 0.019 iCoin 0.0026 Coino 0.0023 Animecoin 0.0000061 Bitcoin 607.25 Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 20, 2014, 07:42:16 PM I checked the model's results earlier today. Yesterday's picks did quite well. Here were the results:
Losers Change 66 Coin -47.5485268630849% ProsperCoin 0% Extremecoin -1.44927536231885% BlackCoin -17.3913043478261% Pesetacoin 0% AVERAGE -13.277821314646% Winners Change Pandacoin (PND) -1.4705882352941% Premine 26.3157894736842% iCoin 30.7692307692308% Coino -34.7826086956522% Animecoin 96.7213114754098% AVERAGE 23.5106269574757% Bitcoin -1.79662412515438% Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 20, 2014, 08:00:04 PM Here's today's picks:
Losers Price Animecoin 0.0000059 LottoCoin 0.000018 CorgiCoin 0.0000014 iCoin 0.0036 Murraycoin 0.0049 Winners Price ContinuumCoin 0.0000038 Grain 0.0000059 66 Coin 1719 CacheCoin 1.43 ORObit 0.0035 Bitcoin 592.13 Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: counter on March 20, 2014, 11:25:14 PM +1 for the OP for putting this thread together I know I appreciate it. Bringing such useful information like this to the masses, it is what the forum needing for sure
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: kdrop22 on March 21, 2014, 01:02:46 AM following...
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 25, 2014, 01:23:07 AM Ok. So, I took the weekend off. I'm back with some picks....
Losers Price Now Grumpycoin 0.0000013 Kittehcoin 0.000058 Gridcoin 0.012 Murraycoin 0.0048 Nyancoin 0.00061 Winners Price Now UltraCoin 0.11 42 Coin 40438 FlorinCoin 0.0035 Vertcoin 1.66 Cryptographic Anomaly 0.88 Bitcoin 581.83 Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: ysbysb2000 on March 26, 2014, 04:37:11 AM BTC >:( :(
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on March 27, 2014, 12:41:31 AM Yesterday's picks didn't do as well as past picks (winner's were down 2% on average when I re-ran the model). I feel like when the model isn't re-ran twice during the day, the picks aren't as accurate.
Today, I reran the model 3x so, with these picks, I'm pretty confident. Let's see how these picks turn out: Here's today's picks: Losers Price Now Pesetacoin 0.032 Marscoin 0.016 ORObit 0.0033 Pxlcoin 0.017 FlappyCoin 0.0000046 Winners Price Now Ixcoin 0.03 Elacoin 0.18 CasinoCoin 0.0016 Joulecoin 0.0023 DigiByte 0.0002 Bitcoin 581.64 Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: PyroClasTix on March 31, 2014, 07:45:11 AM Aww I haven't been on much so haven't been checking in here much.
How's the picks been? Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: Globb0 on March 31, 2014, 07:54:08 AM Interesting :)
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: bitinlet on April 01, 2014, 05:46:55 PM Today, I've decided to post the entire model results (rather than a couple picks), so people get an idea of what's actually occurring behind the scenes. So, the price listed is the price that was modeled. It was obtained from coinmarketcap within the past hour of this post. I ran that last price through the model (using the other variables in the model as well) and out we receive a percentage for each coin. That percent is the predicted percentage gain over the next day. Here's the results:
http://i62.tinypic.com/2r45xeg.png Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: WayToGo on April 01, 2014, 06:37:04 PM Nice thread.Thanks for this.
Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: BTCMS on May 23, 2014, 07:26:18 PM So you've got about 20 days of data? Hows the ROI on average for the winners/losers?
Obviously it is very difficult if not impossible to create a model that can predict a winner with historical data but it is fun and mighty interesting to try nonetheless. Good luck! Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: teamcryptonator on May 30, 2014, 09:54:34 PM have a look at this listing of winners & losers for the 24H / week / month
https://www.cryptonator.com/winners-losers Title: Re: Forecasting Altcoin "Daily Price Change" Winners and Losers Post by: raganius on June 29, 2014, 01:41:35 AM Nice thread. Thank you. Please do your best!
|