Hey guys. Just hoping for some help with my thesis. We're doing an empirical study on how well returns correlate with banks' risk ratings of mutual funds and have some quick detail questions.

1. When it comes to a benchmark index to calculate risk-adjusted performance measures, we were told by our rather confused professor that we should use a few different indices, since our mutual funds differ so greatly (bond funds, asia funds etc).

THE QUESTION IS: do we just create an equally-weighted index consisting of the most important indices for our 36 chosen funds? To me, it seems a bit unscientific with regards to weights etc.

OR - do we compare each fund to the appropriate index? That surely can't be right since the result would be uncomparable, correct?

Any other solutions?


2. Risk-free rate.

We have 8 years of data on the funds and the indices used for creating our own "world index". Is it still okay to use the 3 month treasury bill? Our professor said it was fine, but he was very stressed and I read the following about the problem:

"There can be said to exist two basic conditions for a security to be considered risk-free. The first is that there can be no default risk.
The second condition is that there can be no reinvestment risk. The risk-free rate for a five-year time horizon has to be the expected return on a default-free government zero-coupon bond. If you use the six-month treasury bill rate for example, you face a reinvestment risk of not knowing what the rate will be in six months (Damodaran). "

3. Arithmetic or geometric mean? I'm pretty sure we're supposed to use geometric since we're dealing with percentages over a period of time, right?


Appreciate any help!

This thread may last another 4 minutes.