2 ms·

But do you calibrate p (say through estimation) and then apply the Kelly criterion in your portfolio? I don’t think it is used in this way. It swings too much

by wenc 2y ago

But do you calibrate p (say through estimation) and then apply the Kelly criterion in your portfolio? I don’t think it is used in this way. It swings too much with a given p.

uoaei 2y ago
You calibrate for a reasonable distribution of p and use that to estimate (Monte Carlo, etc.) expected gain, optimizing your investment based on that. With this technique your estimate will probably end up somewhere around the common heuristics.