BETI: Bitcoin Exponential Trend Index and technical analysis

[jl2012]

2015-11-04:

UPDATE

Name of this thread is changed to "BETI: Bitcoin Exponential Trend Index and technical analysis". It will not only include a regular update of the BETI, but also general technical analysis and observations.

BETI = ln (daily VWAP / exponential tread expected price)

For example, on 2015-11-02, the VWAP was 347.89 and the expected price was 1848.06. BETI = ln (347.89/1848.06) = -1.670

Data source and calculation of the expected price are described below

Please consider to donate if you like this service

-------------------
2014-02-17:

Starting from today, I will offer a regularly update of the bitcoin long-term exponential trend.

Method:

Data source: bitcoincharts.com daily volume weighted average price (VWAP)

Reference exchange:
  17 Jul 2010 to 18 Jun 2013: MtGox
  19 Jun 2013 to now: Bitstamp

Dates omitted:
  20 Jun 2011 to 25 Jun 2011 (MtGox closed)
  6 Jan 2015 to 8 Jan 2015 (Bitstamp closed)

Regression model:
  y = ax + b,
  where y = natural logarithm of VWAP; x = days since MtGox inception, 17 Jul 2010 = 0; a = slope; b = intercept

Glossary:
  Rsq: R-square value of the regression model
  Today's expected price: The expected price of today based on the regression model
  Predicted date for today's price: The expected date for today's price based on the regression model
  Days ahead (behind): The difference between today and the predicted date for today's price. A positive value indicates we are currently above the regression and vice versa
  Daily price rank: The rank of today's VWAP among all historical VWAP. 1 means the all-time-high VWAP

------------------------

Date: 16 Feb 2014
VWAP = 625.45
x = 1310
a = 0.006037
b = -1.82098
Rsq = 0.871816
Today's expected price = 440.0484
Predicted date for today's price = 15 Apr 2014
Days ahead = 58
Daily price rank = 84
Predicted date for ATH ($1126) = 26 Jul 2014
------------------------

tl;dr:

ASSUMING the bitcoin price is growing with an exponential trend in long-term:

We expect bitcoin price to grow by about (1 - e^a) = (1 - e^0.006037) = 0.6055% per day
87.1816% (Rsq) of the variation in bitcoin price could be explained solely by time (which is very high)
The long-term "fair" price of today is only 440.0484. We are outpacing the long-term trend.
We expect to see today's price (625.45) on 15 Apr 2014, which is 58 days later
We expect to see ATH (1126) on 26 Jul 2014
Today's price is the 84th highest in the history of bitcoin

-----------------------------



y-axis is ln(price)
Blue line is the daily VWAP
Red line is the expected price of the day. For each day, a regression is fitted with all data of and before that day, so it is not a straight line.
Green line is the current regression line

[jl2012]

Some interesting days:

Date: 9 Jun 2011, where the VWAP ATH at 29.58
VWAP = 29.58
x = 327
a = 0.015305
b = -3.48544
Rsq = 0.903626
The day's expected price = 4.569171
Predicted date for the day's price = 9 Oct 2011
Days ahead = 122
Daily price rank = 1

Date: 18 Nov 2011, where the VWAP bottomed at 2.14
VWAP = 2.14
x = 489
a = 0.012087
b = -2.97171
Rsq = 0.813339
The day's expected price = 18.88924
Predicted date for the day's price = 21 May 2011
Days behind = -180
Daily price rank = 198

[jl2012]

Also, it seems fundamentally odd to consider Mt. Gox in any 'long term' measure of btc considering their present situation.

Please read

[JimboToronto]

it seems fundamentally odd to consider Mt. Gox in any 'long term' measure of btc

While it has recently become almost irrelevant, only Mt Gox has been around long enough to be considered "long term".

What else would you use? Bitstamp? It doesn't even go back as far as the June 2011 bubble.

I'd like to see both Gox and Stamp plotted on the same chart, the one with a tenth of a penny as a base line (as used for Gox) rather than the one starting at a dollar (as used for Stamp). You'd get a truer visual representation that way and the individual could decide when to start disregarding Gox and focusing on Stamp.

[jl2012]

it seems fundamentally odd to consider Mt. Gox in any 'long term' measure of btc

While it has recently become almost irrelevant, only Mt Gox has been around long enough to be considered "long term".

What else would you use? Bitstamp? It doesn't even go back as far as the June 2011 bubble.

I'd like to see both Gox and Stamp plotted on the same chart, the one with a tenth of a penny as a base line (as used for Gox) rather than the one starting at a dollar (as used for Stamp). You'd get a truer visual representation that way and the individual could decide when to start disregarding Gox and focusing on Stamp.

Anyway, I switched to Bitstamp on 19 Jun 2013, when MtGox started having fiat withdrawal problem

[BuildTheFuture]

No visual aid pretty chart? 

[jl2012]

No visual aid pretty chart? 

Maybe I would add it later if I could make it automatic

[JimboToronto]

I switched to Bitstamp on 19 Jun 2013, when MtGox started having fiat withdrawal problem

I think most people made the switch sometime last summer.

It really didn't mean that much to me as I only used Gox for daytarding BTC, never for depositing or withdrawing fiat. I use Virtex as a fiat portal (plus off-exchange purchases) but took advantage of greater volatility and lower tarding fees at Gox for daytarding only.

The closest I come now to either Gox or Stamp is when I buy something with BTC through a service like Bitpay, although the BTC ATM in my neighborhood bases its prices on Stamp plus 5%. That's competitive with the cost of moving money in and out of exchanges, while being much faster and more convenient. It's also possible to buy anonymously with a little care.

Hopefully as time goes on, exchanges will become less and less important for many Bitcoiners.

[JimboToronto]

No visual aid pretty chart?  

Maybe I would add it later if I could make it automatic

+1

That would be excellent, but as I suggested earlier please plot multiple exchanges and use an appropriately small base line.

Edit- Maybe Ritchie (Chartbuddy) could help get you started.

[Alonzo Ewing]

Can you add when we should see major milestones? 
$1000
$2000
$5000
$10000
Etc

[derpinheimer]

Can you add when we should see major milestones? 
$1000 - Tomorrow
$2000 - Friday
$5000 - Tuesday morning CST
$10000 - Tuesday afternoon CST
Etc

yw

[Mythul]

Like the thread, maybe some charts would be nice. People really like to stare at pretty charts that show the moon

Keep up the good work !

[JimboToronto]

I take it that you no longer use Gox, if so what is your #2 or backup exchange preference for the "day trading" ? 

I quit daytrading (actually it was closer to monthtrading) last spring. It wasn't worth the stress and hassle. If I were to use a foreign exchange now it would probably be Bitstamp.

do you pay for the btc at the atm with cash or credit card or multiple ways? 

I don't use credit cards, not that they're an option anyway. No BTC ATM operator would allow payments that could be charged back. It's strictly good old-fashioned non-refundable cash.

The Ontario-made BTM-branded ATM at Bitcoin Decentral on Spadina Ave south of King St. only dispenses BTC in return for CAD $5, $10, and $20 bills fed into its slot, not the other way around, but larger denominations and BTC-to-fiat trades are supposedly in the works.

Anti-money-laundering and KYC requirements are fulfilled through cellphone verification by requiring a cellphone number for the ATM to text a code to, which is then entered into the ATM to verify the transaction. Since most people have their cellphones connected to their identities, this is deemed sufficient. I activated a SIM card for a fully anonymous convenience-store pay-as-you-go cellphone (using a public wifi hotspot through a disposable USB wifi adaptor in a fully anonymous netbook) to maintain secure anonymity. I let the ATM scan the public key QR code and deposit the BTC to a paper wallet that has never been online.

Some ATMs like the world's first Nevada-made Robocoin in Vancouver use palmprint scanning for identity verification. I guess if you want anonymity there, you need to borrow someone's 3-year-old kid or use a silicone mold of an altered palmprint. 

j/k about the fraudulent palmprint of course. I have no reason to seek anonymity except on general principles and proof-of-concept. If you really want anonymous coins it's easy enough just to pay 10% over Virtex ask price to anonymously buy them from someone met through Localbitcoins or a local Bitcoin meetup.

[spiderbrain]

Thanks, great thread.

[JimboToronto]

Thanks, great thread.

+1

Please keep it up jl2012.

[jl2012]

Update:

Date: 23 Feb 2014
VWAP = 619.69
x = 1317
a = 0.006044
b = -1.82413
Rsq = 0.873783
Today's expected price = 461.9526
Predicted date for today's price = 12 Apr 2014
Days ahead = 49
Daily price rank = 90
Predicted date for ATH ($1126) = 25 Jul 2014

(See OP for explanation)

[segeln]

what about graphs?
Visualizing is easier to understand and to catch

[jl2012]

y-axis is ln(price)
Series 1 is the daily VWAP
Series 2 is the expected price of the day. For each day, a regression is fitted with all data of and before that day, so it is not a straight line.

This graph shows that we have experienced a real bear market once only, from 2011 to early 2013. The mid-2013 bear market was basically following the mean

[segeln]

y-axis is ln(price)
Series 1 is the daily VWAP
Series 2 is the expected price of the day. For each day, a regression is fitted with all data of and before that day, so it is not a straight line.

This graph shows that we have experienced a real bear market once only, from 2011 to early 2013. The mid-2013 bear market was basically following the mean

great,thanks

so,now we are still in a bull-Phase,right? Edging the bear-trend
but the data go only to 17/1/2014

[jl2012]

y-axis is ln(price)
Series 1 is the daily VWAP
Series 2 is the expected price of the day. For each day, a regression is fitted with all data of and before that day, so it is not a straight line.

This graph shows that we have experienced a real bear market once only, from 2011 to early 2013. The mid-2013 bear market was basically following the mean

great,thanks

so,now we are still in a bull-Phase,right?
but the data go only to 17/1/2014

The last data point is yesterday, just not shown on the x-axis

With this graph, we may even say we are still in a big bull-phase starting from Mar 2013. The price is still >10x than 1 year before, anyway.