Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Is Anyone Using The Hurst Exponent?
Peter Cawthron
edited May 6, 2013

Hello All,

The Hurst Exponent can be calculated for a time series as a value for 0 to 1. A value of 0.5 says the time series is a random walk, < 0.5 that it is mean-reverting, and > 0.5 that it is trending. I've over-simplified a bit so see these references:

'Rescaled Range Analysis: A Method for Detecting Persistence, Randomness, or Mean Reversion in Financial Markets' http://blogs.cfainstitute.org/investor/2013/01/30/rescaled-range-analysis-a-method-for-detecting-persistence-randomness-or-mean-reversion-in-financial-markets/

'Some comments on Hurst exponent and the long memory processes on
capital markets' http://www.ual.es/~jgarcia/index_archivos/HURST.pdf

I've copied some code from the PyEEG project http://code.google.com/p/pyeeg/ but initially I am getting Hurst Exponent values for SPY and for individual stocks that are around 0.7. In both references above the S&P 500 has a calculated Hurst Exponent of 0.49 - 0.5.

What am I doing wrong? And is there a trading method here?

Regards,

Peter

5 responses
Damián Avila
May 6, 2013

Hi,

If I remember correctly, there are a lot of methodologies to "estimate" the Hurst Exponent. You have some of them listed here: https://www1.ethz.ch/mosaic/research/docs/Racine2011.pdf‎
I have not seen your code, but if the code is right, maybe you have a "biased" Hurst Exponent because of the sample data you are using...
You have a comparative article about this problem: www.foibg.com/ijitk/ijitk-vol05/ijitk05-4-p05.pdf‎

BTW, as far as I know (but I am new here) there is not an example using it here...

Cheers.

Damián.

Peter Cawthron
May 6, 2013

Hello Damián,

Thanks for the links. They work but only if I delete '%E2%80%8E' from the end i.e.:

https://www1.ethz.ch/mosaic/research/docs/Racine2011.pdf

http://www.foibg.com/ijitk/ijitk-vol05/ijitk05-4-p05.pdf

Regards,

Peter

(If anyone is curious this is due to the UTF-8 left-to-right mark (LRM). See: http://en.wikipedia.org/wiki/Left-to-right_mark )

Zoe Miao
Jan 20, 2016

Hi Peter, usually use logarithm return of the stock prices as a time series to feed into the model. I saw that you're using the prices directly. This might be the problem.

Mat James
Apr 25, 2016

I've dived into some Hurst Academic papers, that's quite brilliant.

Does anyone has good results by filtering market ?

At which timeframe does it mostly apply, I think it's mostly on daily/weekly isn't?

Carter Pelot
Aug 2, 2016

Peter - I agree with Zao. From my understanding, you should be using the log return not prices directly.

About Quantopian
Careers
Community
Events
Documentation
Academia
Lectures
Status
Facebook
Twitter
LinkedIn
YouTube
Terms of Use
Privacy Policy

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by Quantopian.

In addition, the material offers no opinion with respect to the suitability of any security or specific investment. No information contained herein should be regarded as a suggestion to engage in or refrain from any investment-related course of action as none of Quantopian nor any of its affiliates is undertaking to provide investment advice, act as an adviser to any plan or entity subject to the Employee Retirement Income Security Act of 1974, as amended, individual retirement account or individual retirement annuity, or give advice in a fiduciary capacity with respect to the materials presented herein. If you are an individual retirement or other investor, contact your financial advisor or other fiduciary unrelated to Quantopian about whether any given investment idea, strategy, product or service described herein may be appropriate for your circumstances. All investments involve risk, including loss of principal. Quantopian makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances.