Comment by smabie
4 years ago
I find both mean variance and Kelly to be very poor in practice due to the dependence on the expected return term. Like, if I knew that, I wouldn't be wasting my time with all this math! (half joking)
4 years ago
I find both mean variance and Kelly to be very poor in practice due to the dependence on the expected return term. Like, if I knew that, I wouldn't be wasting my time with all this math! (half joking)
Kelly does not depend on the expected return -- it depends on the joint distribution of outcomes. That is a big difference!
Accidentally thinking that "E log X" and "log E X" are the same thing is a common mistake, but Jensen's inequality tells us it can be a costly one.
Of course, your general point still stands: if we only knew the joint distribution of outcomes the battle would be over already.