Back to Community
ideas for algorithms

A recent reddit post on the algotrading sub-reddit is about a survey paper of 100+ prediction techniques, which could be a great source of ideas for algorithms to implement here. Here is the paper: http://www.scribd.com/doc/112855946/Surveying-Stock-Market-Forecasting-Techniques-Part-II-Soft-Computing-Methods-2009

And I found it here: http://www.reddit.com/r/algotrading/comments/130ct0/surveying_stock_market_forecasting_techniques/

Disclaimer

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.

2 responses

Hello Fawce,

Your post reminded me of a site I'd bookmarked awhile back when I saw it in the news...not sure if it'd be useful in developing trading algorithms, but I thought I'd share it since it is a recent innovation. The site, along with some related links:

http://www.exploredata.net

http://www.sciencemag.org/content/334/6062/1518

http://www.uvm.edu/~cmplxsys/newsevents/pdfs/2012/reshef-correlation-science-2011.pdf

A "Python wrapper" is available here:

https://github.com/ajmazurie/xstats.MINE

@Grant, fascinating paper! I had read about the two brothers in the popular press, but this is the first time I've read anything about the MINE/MIC algorithm. They provide an intuitive explanation at the beginning of the article (third link in your list).

The benefit of the technique is looking for fit, regardless of the relationship between variables. To be of big benefit we would need data where we could imagine finding non-linear relationships - otherwise I would imagine it is computationally cheaper to just test a linear regression (though I don't know if that is true).