Information Theory and Portfolio Management
Information Theory has come a long way since Shannon’s Channel Coding Theorem. The beauty of Information Theory in giving a totally new understanding of the world as it is, is humbling. I have not done much reading on Fractals[ Most of my education were because of books which were not in my course 
, no wonder my dad always had to hide his face, whenever the sem exams ended 
], but I distinctly remember a warning on a book of Fractals which also served as a “hook”. It read, “Beware before turning the page, for your vision of nature’s beauty can change forever”.
It was intriguing, and though it was a clever “hook”, I always felt the overlap of Information Theory with so vast and diverse playing field was awe-inspiring.
In the book, “Elements of Information Theory”, Thomas Cover there is an image which shows, the overlap of IT with almost all the modern sciences right from computational complexity, to stock markets to digital communication.
If amongst my reader, I do have anybody who nurtures a slight academic bent, he might find this paper[link below] helpful in building up his information theoretic approach to capital markets. The paper is a student project and is formed on the basis of “Elements of Information Theory” , the book which I just mentioned. Introduced from bottom up, you might like to give it a read.
Sorry about this off-topic question. Wanted to know if you are aware of any stock ranking (not ROC or Relative Strength based) papers that I can read about.
Hi Prashanth, Stock Ranking Papers as in?
Thanks for the reply. To be honest, I don’t know for sure what I am looking for. I am working on building a stock trading strategy that shall pick stocks based upon some criteria which I still am not sure off.
I have tried Relative Strength based picking (Buy Strong and Rotate on either weekly or fortnightly basis), but results aren’t as good as I was expecting. Some sectors show strong co-relation while others haven’t. If results do not give a edge, what good would it be compared to just buying the hot stocks or even simpler, just tossing the coin where if done for a long time, one has a probability of 50% success. Mix some Money Management and that itself may outperform my strategy
This has moved me to see what kind of strategies (without Fundamental filters) can be thought of. I have read quite a bit and searched even more, but out in the open, there seems to be no idea providers
. Hence looking for papers which have done studies as to whether on a longer run, there is any way to extract info from RS.
Hmm, there seems to be a lot many more “unknowns” in the equation.
Your selection of the filter which will pick stocks largely depends on the strategy you are employing. Trend Following will have different master-filter[if I am allowed to use it
] when compared to trend fading, or stat. arb for that matter.
You might like to ping me at my email, if you like to talk.