All of Paul Lang's Comments + Replies

Last time I checked, E[ln(X)] = ln(G(X)) where G(X) is the geometric expectation of X

Not an expert, but to me this math checks out.

First of all, thank you very much for this thought provoking post. I'm not sure if I've arrived at the right conclusions here, but it seems that

we have made a $12 profit and the insurer has made a $37 profit

is not technically correct. The reason is that you cannot stick units into a logarithmic function. So what you get out on the left hand site is unitless.

But there is another way to think about this. As stated on Wikipedia, "In probability theory, the Kelly criterion (or Kelly strategy or Kelly bet) is a formula for sizing a sequence of bets by maximizin... (read more)