16th Oct 07

Dome (at the moment, my suggestions to you and to Muhannad are the same),

I think a really good start would be to read a selection of the papers from these two researchers: Michael Dempster, Cambridge, and Andy Lo, MIT. These were suggested by my colleague Professor Peter Ross, who is now mainly out of academia and working almost full time in the finance industry using machine learning methods.

Lo's papers are here: http://web.mit.edu/alo/www/articles.html -- start with:

• Computational Challenges of Financial Engineering
• Finance: A Selective Survey
• Foundations of Technical Analysis: Computational Algorithms, Statistical Inference, and Empirical Implementation

The third one is quite long, and will be very useful if we do something along the lines of `technical analysis', but this is not the only approach in finance. Well worth reading and understanding.

Dempster's papers are here: http://mahd-pc.jbs.cam.ac.uk/publish/

Have a look around these papers -- what your PhD ends up doing may or may not be closely related to Dempster's work. E.g. this describes a typical (though rather good) system that applies to finance a machine learning approach from my field of interest:

http://mahd-pc.jbs.cam.ac.uk/archive/PAPERS/2004/WP18.pdf

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