Bitcoin Forum
June 27, 2019, 01:43:12 PM *
News: Latest Bitcoin Core release: 0.18.0 [Torrent] (New!)
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: A Predictive Model for the Growth of Bitcoin Illustrated in Pricing Data Pattern  (Read 3714 times)
jr3951
Newbie
*
Offline Offline

Activity: 59
Merit: 0


View Profile
February 16, 2014, 07:02:35 AM
Last edit: February 16, 2014, 07:43:29 AM by jr3951
 #1

FOREWORD-
I was just looking at prices on coinbase today and noticed a certain pattern in how it had moved, I compared it to a previous pattern and that to another one and noticed a definite trend, so I stayed up all night and made this… I hope you enjoy and find it as informative as I did making it, if it interests you feel free to add on the model and make predictions yourself.

For the sake of separating trends I classified each stage of bitcoin growth, when I started I picked the recent peak and the one before that, and so on, all data is from the coinbase chart, so other markets will have varying data but generally the same trends


http://i59.tinypic.com/21478s0.jpg


Incubation Period- “penny stock” growth, resulting in first mainstream attention, crescendo of growth at end
Period Dates- Oct 16, 2010 to Jun 8, 2011
Total 236 Days
Period Growth-29,500%


http://i61.tinypic.com/2i6gdhl.jpg


Birth Period- Large crash, Stabilization period, dollar stock growth, Predictable Pattern begins taking form
Period 1 Dates-
Jun 8, 2011 to Jan 8, 2012
Total: 215 Days
Period 1 Decline- 316%

Period 2 Dates-
Jan 8, 2012 to Aug 16, 2012
Total 222 Days
Period 2 Growth- 89.87%
Period 2 Crash-40.74%
Period 2 Trough A- 11 Days, 24.62% Growth
Period 2 Trough B- 56 Days, 02.90% Growth



http://i61.tinypic.com/2041ky.jpg


http://i60.tinypic.com/9qzo7d.jpg

Adolescence/ Rapid Growth Period- sustained growth by large factors, each with 2 troughs and period of stabilization or steady growth

Period 1 Duration:
Aug 19, 2012 to Apr 9, 2013
Total: 233 Days
Period 1 Growth- 1,603%
Period 1 Crash- 70.43%
Period 1 Trough A – 17 Days, Growth- 44.11%
Period 1 Trough B- 64 Days, Decline- 31.77%
Period 1 Trough C- 88 days, Growth- 49.28
BP2 to AP1 Bottoms- 750% Growth

Period 2 Duration:
Apr 9, 2013 to Dec 4, 2013
Total 240 days
Period 2 Growth- 398%
Period 2 Crash- 39.49%, second dip total -54.48%
Period 2 Trough A- 11 Days, Decline- 24.78%
Period 2 Trough B- 57 Days, Growth- 14.55%
Total Period 2 Trough 68 Days, Decline 13.83%
Period 1 Bottom to Period 2 crash bottom- 920% Growth

Applying This Model to the Future:
Using patterns and guessing we can predict:
General growth from here until the next plateau, the deepest trough is usually the 1st or 2nd, we will probably not go lower than the deepest trough already sustained at this level.
-   A possible 3rd trough in early May but with a large net growth when compared to the second

The next Bitcoin market peak will be Aug 1, 2014 given 240 days
-   There have been shorter cycles, a 215 Day Model (Shortest Cycle), puts an earliest date of July 7, 2014.
-   Cycles have been trending increased durations, slightly over 240 days could be expected given momentum.
-   The preceding Crash has been trending downward in overall effect, a crash totaling up to *only* 45% off the value at the deepest trough, even just 20-30% given the right outside circumstances
-   The total increase of the future high from the last peak has been decreasing, though I personally think the degree is just an illusion from a slightly skewed peak of 230$ (which shows in the large crash), I think the very conservative numbers say 3,000 USD, several other trends suggest 10,000 USD at the next peak
After the immediate crash a secondary high is expected between an 11-17 day timeframe, followed by another trough
-   A 700-1000 or more % growth might be expected between the Dec 18th Trough and the future deepest trough, leaving that trough at over possibly 5,000 USD at the lowest point.

Things to consider:
-   August 1st, 2014 could be the day they release a bitcoin killer altcoin, obviously by no means does any voodoo guesses I do have any effect on what the market ultimately does
-   These are just my interpretations of the numbers, feel free to draw your own patterns and conclusions
-   Labeling different stages of the bitcoin market cycles helps differentiate the patterns, but also means patterns can of course radically change
-   -Eventually one would expect the growth cycles to start getting farther between ending in a “maturity” phase of slower sustained growth
-   At its core I think this is a good case study into how burgeoning technologies are developed in value over time at a rate almost eerily similar to Moore’s law (every 8 months bitcoin will have increased or decreased based on certain factors given momentum, this value currently seems to be around x10)

Donations of Money Always Appreciated Smiley
1KK4dtEf3J87ccdWMZNC2VPcimUs9Aaidq
1561642992
Hero Member
*
Offline Offline

Posts: 1561642992

View Profile Personal Message (Offline)

Ignore
1561642992
Reply with quote  #2

1561642992
Report to moderator
1561642992
Hero Member
*
Offline Offline

Posts: 1561642992

View Profile Personal Message (Offline)

Ignore
1561642992
Reply with quote  #2

1561642992
Report to moderator
1561642992
Hero Member
*
Offline Offline

Posts: 1561642992

View Profile Personal Message (Offline)

Ignore
1561642992
Reply with quote  #2

1561642992
Report to moderator
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1561642992
Hero Member
*
Offline Offline

Posts: 1561642992

View Profile Personal Message (Offline)

Ignore
1561642992
Reply with quote  #2

1561642992
Report to moderator
kolev
Member
**
Offline Offline

Activity: 94
Merit: 10

My name is Nikolay and am a webaholic.


View Profile WWW
February 16, 2014, 07:12:22 AM
 #2

FOREWORD-
I was just looking at prices on coinbase today and noticed a certain pattern in how it had moved, I compared it to a previous pattern and that to another one and noticed a definite trend, so I stayed up all night and made this… I hope you enjoy and find it as informative as I did making it, if it interests you feel free to add on the model and make predictions yourself.

For the sake of separating trends I classified each stage of bitcoin growth, when I started I picked the recent peak and the one before that, and so on, all data is from the coinbase chart, so other markets will have varying data but generally the same trends





Incubation Period- “penny stock” growth, resulting in first mainstream attention, crescendo of growth at end
Period Dates- Oct 16, 2010 to Jun 8, 2011
Total 236 Days
Period Growth-29,500%





Birth Period- Large crash, Stabilization period, dollar stock growth, Predictable Pattern begins taking form
Period 1 Dates-
Jun 8, 2011 to Jan 8, 2012
Total: 215 Days
Period 1 Decline- 316%

Period 2 Dates-
Jan 8, 2012 to Aug 16, 2012
Total 222 Days
Period 2 Growth- 89.87%
Period 2 Crash-40.74%
Period 2 Trough A- 11 Days, 24.62% Growth
Period 2 Trough B- 56 Days, 02.90% Growth








Adolescence/ Rapid Growth Period- sustained growth by large factors, each with double bottom cycles and period of stabilization

Period 1 Duration:
Aug 16, 2012 to Apr 9, 2013
Total: 237 Days
Period 1 Growth- 1,603%
Period 1 Crash- 70.43%
Period 1 Trough A – 17 Days, Growth- 44.11%
Period 1 Trough B- 64 Days, Decline- 31.77%
Period 1 Trough C- 88 days, Growth- 49.28
BP2 to AP1 Bottoms- 750% Growth

Period 2 Duration:
Apr 9, 2013 to Dec 4, 2013
Total 240 days
Period 2 Growth- 398%
Period 2 Crash- 39.49%, second dip total -54.48%
Period 2 Trough A- 11 Days, Decline- 24.78%
Period 2 Trough B- 57 Days, Growth- 14.55%
Total Period 2 Trough 68 Days, Decline 13.83%
Period 1 Bottom to Period 2 crash bottom- 920% Growth

Applying This Model to the Future:
Using patterns and guessing we can predict:
General growth from here until the next plateau, the deepest trough is usually the 1st or 2nd, we will probably not go lower than the deepest trough already sustained at this level.
-   A possible 3rd trough in early May but with a large net growth when compared to the second

The next Bitcoin market peak will be Aug 1, 2014 given 240 days
-   There have been shorter cycles, a 215 Day Model (Shortest Cycle), puts an earliest date of July 7, 2014.
-   Cycles have been trending increased durations, slightly over 240 days could be expected given momentum.
-   The preceding Crash has been trending downward in overall effect, a crash totaling up to *only* 45% off the value at the deepest trough, even just 20-30% given the right outside circumstances
-   The total increase of the future high from the last peak has been decreasing, though I personally think the degree is just an illusion from a slightly skewed peak of 230$ (which shows in the large crash), I think the very conservative numbers say 3,000 USD, several other trends suggest 10,000 USD at the next peak
After the immediate crash a secondary high is expected between an 11-17 day timeframe, followed by another trough
-   A 700-1000 or more % growth might be expected between the Dec 18th Trough and the future deepest trough, leaving that trough at over possibly 5,000 USD at the lowest point.

Things to consider:
-   August 1st, 2014 could be the day they release a bitcoin killer altcoin, obviously by no means does any voodoo guesses I do have any effect on what the market ultimately does
-   These are just my interpretations of the numbers, feel free to draw your own patterns and conclusions
-   Labeling different stages of the bitcoin market cycles helps differentiate the patterns, but also means patterns can of course radically change
-   -Eventually one would expect the growth cycles to start getting farther between ending in a “maturity” phase of slower sustained growth
-   At its core I think this is a good case study into how burgeoning technologies are developed in value over time at a rate almost eerily similar to Moore’s law (every 8 months bitcoin will have increased or decreased based on certain factors given momentum, this value currently seems to be around x10)

Donations of Money Always Appreciated Smiley
1KK4dtEf3J87ccdWMZNC2VPcimUs9Aaidq
I've always been amused when people try to find order in chaos.

Bitrated user: nikolay.
jr3951
Newbie
*
Offline Offline

Activity: 59
Merit: 0


View Profile
February 16, 2014, 07:19:43 AM
 #3

obviously by no means does any voodoo guesses I do have any effect on what the market ultimately does

I've always been amused when people try to find order in chaos.
odolvlobo
Legendary
*
Offline Offline

Activity: 2534
Merit: 1332



View Profile
February 16, 2014, 07:24:34 AM
 #4

Total fantasy. You believe that history will repeat itself when you have absolutely no evidence to support that belief. And even if it does, you have no way to know what the time scale might be.

Buy stuff on Amazon at a discount with bitcoins or convert Amazon points to bitcoins: Purse.io
Join an anti-signature campaign: Click ignore on the members of signature campaigns.
jr3951
Newbie
*
Offline Offline

Activity: 59
Merit: 0


View Profile
February 16, 2014, 07:26:18 AM
Last edit: February 16, 2014, 09:33:29 PM by jr3951
 #5

Did you read? The distance between each high spot in the market is almost exactly 8 months, 5 times in a row

I understand the Gamblers Fallacy, and Im quick to make that distinction

But 5 times in a row is in the very least a trend sir

Edited
odolvlobo
Legendary
*
Offline Offline

Activity: 2534
Merit: 1332



View Profile
February 16, 2014, 07:48:08 AM
 #6

Did you read? The distance between each high spot in the market is almost exactly 8 months, 7 times in a row
But 7 times in a row is in the very least a trend sir

First, you show 5 peaks, not 7, but that is irrelevant.

You have demonstrated what is called "selection bias". You picked peaks that fit your model. There are many peaks that you did not pick, presumably because they didn't fit. If you pick all the peaks (if that is even possible without some kind of bias), your model would crumble.

Buy stuff on Amazon at a discount with bitcoins or convert Amazon points to bitcoins: Purse.io
Join an anti-signature campaign: Click ignore on the members of signature campaigns.
jr3951
Newbie
*
Offline Offline

Activity: 59
Merit: 0


View Profile
February 16, 2014, 07:51:24 AM
Last edit: February 16, 2014, 08:26:15 AM by jr3951
 #7

You have demonstrated what is called "selection bias". You picked peaks that fit your model. There are many peaks that you did not pick, presumably because they didn't fit. If you picked all the peaks (if that were even possible without some kind of bias), your model would crumble.

Check the Graphs again, I only pick the highest point peak (singular) in an 8 month time segment, and it has always landed within a margin of 25 days (215-240), except the "bitcoin hits 10 cents" thing, and i did miscount 1 (I am somewhat tired)

but 5 times in a row still constitutes a trend

Not to mention 3 times in a row we have seen a double trough trend with an 11-17 day high period between, that is point immediately after crash, then next lowest point after
odolvlobo
Legendary
*
Offline Offline

Activity: 2534
Merit: 1332



View Profile
February 16, 2014, 12:17:39 PM
 #8

Check the Graphs again, I only pick the highest point peak (singular) in an 8 month time segment, and it has always landed within a margin of 25 days (215-240),

Did you read? The distance between each high spot in the market is almost exactly 8 months

You have proven my point. Look at the lines highlighted above. First, you chose peaks 8 months apart. Then, you claimed that there is a pattern. The pattern is really in your choosing. You chose the ones that were 8 months apart. There are others that are not 8 months apart, such as the September 2011, May 2013 and the January 2014 peaks, but you did not choose them.

Buy stuff on Amazon at a discount with bitcoins or convert Amazon points to bitcoins: Purse.io
Join an anti-signature campaign: Click ignore on the members of signature campaigns.
cdog
Hero Member
*****
Offline Offline

Activity: 812
Merit: 500


View Profile
February 16, 2014, 02:06:38 PM
 #9

But its a trend bro. I looked at the chaaaaarrrts. Lets make moooonnneeeeyyyy



jr3951
Newbie
*
Offline Offline

Activity: 59
Merit: 0


View Profile
February 16, 2014, 08:01:40 PM
Last edit: February 16, 2014, 09:34:14 PM by jr3951
 #10

Check the Graphs again, I only pick the highest point peak (singular) in an 8 month time segment, and it has always landed within a margin of 25 days (215-240),

Did you read? The distance between each high spot in the market is almost exactly 8 months

You have proven my point. Look at the lines highlighted above. First, you chose peaks 8 months apart. Then, you claimed that there is a pattern. The pattern is really in your choosing. You chose the ones that were 8 months apart. There are others that are not 8 months apart, such as the September 2011, May 2013 and the January 2014 peaks, but you did not choose them.

Because they were not the highest peaks in their respective 8 month time periods, they were also not the peaks that "kicked off" the next plateau, it is irrefutable to suggest at least those 3 specific peaks are not almost exactly 240 days apart each, if you really need Ill draw more charts and show you like a child

Regardless, the cool thing about a predictive model is, we can check in 6 months and see if I'm right
bb113
Hero Member
*****
Offline Offline

Activity: 728
Merit: 500


View Profile
February 16, 2014, 11:01:51 PM
 #11

Cool.

People don't understand what chaos is, its sensitivity to initial conditions that yields prediction impossible because of the butterfly effect. Chaos is posed as deterministc, only the approximation of present conditions does not equal the approximation of their future states. But bitcoin is not a deterministic chaotic system, its indetermined because of its self-referentiality, yet linear, because it has a positive feedback loop. I think it will always be volatile (because, there is no possible equilibirum if the value is more than 0) but its area of volatility will change upwards with its spread, and can be rationaly speculated upon by investigating trends of spreading and the conditions of spreading.

That said the patterns on graphs tell us nothing, they can only serve as a signs to investigate. I'm sure you are right when it comes to the first phase, thats the reason it always pays off to make new coins.


Can you give links or more detail as to why feedback loops are incompatible with deterministic systems?
nwbitcoin
Sr. Member
****
Offline Offline

Activity: 294
Merit: 250


You are a geek if you are too early to the party!


View Profile WWW
February 17, 2014, 12:58:44 PM
 #12

8 month loop is interesting but it doesn't take elements such as volume of trading or factors such as where the data is from.

one of the big problems for bitcoin figures is that it used to be about mtgox,but now the bitcoin price is really free market so the peaks won't match!

Posted From bitcointalk.org Android App

*Image Removed*
I use Localbitcoins to sell bitcoins for GBP by bank transfer!
jr3951
Newbie
*
Offline Offline

Activity: 59
Merit: 0


View Profile
February 17, 2014, 09:56:32 PM
Last edit: February 18, 2014, 02:52:57 AM by jr3951
 #13

8 month loop is interesting but it doesn't take elements such as volume of trading or factors such as where the data is from.

one of the big problems for bitcoin figures is that it used to be about mtgox,but now the bitcoin price is really free market so the peaks won't match!

Posted From bitcointalk.org Android App

The peaks should match across exchanges, primarily the large peaks every 8 months,maybe not the exact dates of the high's across but certainly the relative price fluxing, because of arbitrage...

as to exactly why every 8 months we see exponential price increases/ 8 month highs like clockwork is certainly up for debate though, at least beyond a broad moore's law comparison

Edit: Also, I noticed the patterns on coinBASE originally, the charts are from coinDESK
Pages: [1]
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!