332 reputation
15
bio website
location New York, NY
age
visits member for 2 years
seen Nov 22 at 16:02

I work at a quant fund in NY.

I'm generally interested in statistical learning and optimization.


Jun
12
comment Optimal trading strategy in toy world of simple Hidden Markov model with Gaussians
Richard- I didn't write $E\{...\}$ for any distributions. Also, I don't understand what your point is RE: $\mu_0 = \mu_1$. In that case the solution is a constant, but this is just a special case. In any event, since the OP is trading based on an HMM, it is quite unlikely that $\mu_0 = \mu_1$ here.
Jun
12
comment Optimal trading strategy in toy world of simple Hidden Markov model with Gaussians
Added detailed proof of Sharpe ratio <> mean-variance equivalence. This is a standard result though.
Jun
12
comment Optimal trading strategy in toy world of simple Hidden Markov model with Gaussians
Richard- The forward backward algorithm is exact, so no.
Jun
10
comment Why shrink the covariance matrix?
Positive definite implies non-singular
Jun
9
comment How do I find the most diversified portfolio, or least correlated subset, of stocks?
This answer doesn't really address the question of selecting 5 stocks. You state "If you wish, you can...", but don't really justify your suggestion.
Dec
9
comment George Soros models
I very much agree that game theory is the relevant discipline. In particular, the subfield of learning in games, also called evolutionary game theory.