This indicator is based on a talk given by Kris Kaufman of Parallax Financial Research. The talk was "Jan 2013 MTA Presentation on ExtremeHurst" that can be found at http://www.pfr.com/pfr/articles.htm
Based on the Chaos Theory, the Fractal Market Hypothesis is an alternative to the Efficient Market Hypothesis. It assumes that each trader may interpret trading information in different ways and at different times (The Efficient Market Hypothesis assumes that all information is reflected in prices), that investors are not rational and influenced, in their trading decisions, by what has happened in the past and also by their recent experiences. It also assumes that price changes are not normally distributed (The best example is that usually prices fall much faster than they rise).
The Hurst exponent, also called "H" or "the index of dependence", is part of the Chaos Theory. It is a measure of the predictability of a time series.
A value of H=0.5 means that the market or security follows a true random walk (The stock or market is unpredictable), while a value between 0 and 0.5 indicates that there is a negative autocorrelation
(A decrease in the stock or market is likely to be followed by an increase),
and a value between 0.5 and 1 indicates that there is a positive autocorrelation
(An increase in the stock or market is likely to be followed by an another increase).
The more the Hurst exponent is far from the 0.5 level, the more likely the underlying time-series is predictable.
Comments
thanks
watch this => http://www.pfr.com/files/MTA_Presentation.wmv
Based on the Chaos Theory, the Fractal Market Hypothesis is an alternative to the Efficient Market Hypothesis. It assumes that each trader may interpret trading information in different ways and at different times (The Efficient Market Hypothesis assumes that all information is reflected in prices), that investors are not rational and influenced, in their trading decisions, by what has happened in the past and also by their recent experiences. It also assumes that price changes are not normally distributed (The best example is that usually prices fall much faster than they rise).
The Hurst exponent, also called "H" or "the index of dependence", is part of the Chaos Theory. It is a measure of the predictability of a time series.
A value of H=0.5 means that the market or security follows a true random walk (The stock or market is unpredictable), while a value between 0 and 0.5 indicates that there is a negative autocorrelation
(A decrease in the stock or market is likely to be followed by an increase),
and a value between 0.5 and 1 indicates that there is a positive autocorrelation
(An increase in the stock or market is likely to be followed by an another increase).
The more the Hurst exponent is far from the 0.5 level, the more likely the underlying time-series is predictable.
Thanks jaguar1637 for the explanation on the Hurst exponent.
Some people says about Hurst exponent :
- if you want to use it, you have to take all datas in H1 or H4 from one year, and apply it on a chart
IHMO, as for iVAR and FGDI usage, I would remind you to use it w/ PERIOD_M1. So, you will need much less datas