Jul

22

Probability matching is an interesting phenomena, a bit subtle as to trading.

Probability matching is a decision strategy in which predictions of class membership are proportional to the class base rates. Thus, if in the training set positive examples are observed 60% of the time, and negative examples are observed 40% of the time, then the observer using a probability-matching strategy will predict (for unlabeled examples) a class label of "positive" on 60% of instances, and a class label of "negative" on 40% of instances.

Andrew Lo takes a very deep dive into probability matching:

Evolutionary Foundations of Economic Behavior, Bounded Rationality, and Intelligence
Andrew W. Lo, Massachusetts Institute of Technology
Institute for Pure and Applied Mathematics, UCLA May 19, 2015


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search