I have a data set which consists of > 500 hedge funds, their historical monthly returns, and their benchmark (index) monthly returns. The number of data points (# of monthly returns) differs from a fund to fund (can be as low as 10 and high as 200).

I am wondering if there are some machine learning, advanced statistics, or other quantitative techniques to identify "good" and "bad" funds. I've been googling and I can only find basic statistics such as sharpe ratio, information ratio, etc and compare the statistics between fund and index.

I am not trying to predict the funds' returns. I am only interested in evaluating their returns and risk, using quantitative techniques.

  • $\begingroup$ I guess you need to specify your goal in a more formal way. If you want to classify hedge funds - you need to have a heuristic that says which one is good or bad. Also you can ask an expert to label the sample. $\endgroup$ – Daniel Chepenko Sep 14 '18 at 9:14
  • $\begingroup$ As you have quite small dataset have a look at SVM classifier. $\endgroup$ – Daniel Chepenko Sep 14 '18 at 9:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.