Search
× Search
Saturday, December 28, 2024

Archived Discussions

Recent member discussions

The Algorithmic Traders' Association prides itself on providing a forum for the publication and dissemination of its members' white papers, research, reflections, works in progress, and other contributions. Please Note that archive searches and some of our members' publications are reserved for members only, so please log in or sign up to gain the most from our members' contributions.

Fight overfitting with Turing

photo

 Marcello Calamai, Technology Advisor

 Sunday, December 11, 2016

Trying to apply numerical methods to get alpha, overfitting is our worst enemy. I'd like to hear your opinion about using Kolmogorov complexity to evaluate overfitting. Algorhitmic complexity can be expressed by the dimension of the table of rules of a Turing machine. So, we could evolve a TM starting with a very short table and see how our trading system fits the fitness function. Then, we could gradually enlarge the table until it fits the function to a level that we consider reasonable. This level should express the intrinsic algorithmic complexity of our problem. So, we might assume that every larger table will overfit the parameters.


Print

Please login or register to post comments.

TRADING FUTURES AND OPTIONS INVOLVES SUBSTANTIAL RISK OF LOSS AND IS NOT SUITABLE FOR ALL INVESTORS
Terms Of UsePrivacy StatementCopyright 2018 Algorithmic Traders Association