Quote from: Iris_Colt on September 05, 2018, 07:00:38 AM

Quote from: il_minatore on September 04, 2018, 03:28:03 PM

Hi,
how did you plot the last chart? thanks

Following user jl2012's instructions from their first post, etc.

tl;dr:
Mt. Gox, CSV data available from Saturday, 17 July 2010 23:09:17 to Tuesday, 25 February 2014 1:59:06, implicitly removing from 20 Jun 2011 to 25 Jun 2011 (MtGox was closed).
Bitstamp, CSV data available from Tuesday, 13 September 2011 13:53:36 to Today, implicitly removing from 6 Jan 2015 to 8 Jan 2015 (Bitstamp was closed).
VWAP: daily Volume Weighted Average Price (from bitcoincharts.com), i.e. when computing the average for a given day, the price is being multiplied by the volume of bitcoin traded at that price.
Regression model (computed in C with gsl_fit_linear(): y = ax + b, where y = natural logarithm of VWAP; x = days since MtGox inception, i.e. 17 Jul 2010 = 0; a = slope; b = intercept
Today's exponential tread expected price: the expected price of today based on the regression model, i.e. = ax + b
BETI: Bitcoin Exponential Trend Index. BETI = ln (daily VWAP / exponential tread expected price) [For example, on Sep 2 2018, the VWAP was $7244.62 and the expected price was $15,306.94. BETI = ln ($7244.62/$15,306.94) = -0.748048]

for the last few years, it would be better to use the sum of the shares of more exchange