Describe the five ways sharpe ratio can be gamed by hedge fund managers:
illiquid stocks skew it down long holding period skew it down I can only think of two!
- standard deviation understates the hedge fund downside risk since returns are negatively skewed -longer time periods will increase the ratio by the square root of time -survivorship bias = portfolios that do poorly drop out, which raises average returns
I’m asking about gaming in particular- hint, first one is: 1. Compounding the monthly returns but calculating the standard deviation from the not compounded monthly returns. Since sharpe is all over level 3, this could be an interesting way they test it- who knows the other 4?
-5 for ilvino
Second hint, it’s one of the EOCQ
Sharpe Ratio is over rated. I don’t think it will show up on the exam 
leaving out extreme results - affecst the std deviation and sharpe ratio
+1 dyslexic be back in 20 min, gotta feed my baby
- Lengthening the measurement interval. This will result in a lower estimate of vol. 2. Compounding monthly returns but calculating the std dev from the not compounded monthly returns. 3. writing out of the money puts and calls on a portfolio. This increases returns from collecting an option premium without paying off (up to several years). Strategies that take on default risk, liquidity risk, have same ability to upwardly bias Sharpe. 4. Smoothing of returns with derivatives structures, infrequent mark to market. 5. Getting rid of extreme returns (best and worst returns each year) using a total return swap or directly with options.
Extending measurement period Use of derivatives - selling out of money calls etc. Use of smoothed data
can someone please explain in laymens terms what “upward bias” and “downward bias” means? I’m looking at the question now that inspired this thread by smokin’. it says "discuss situations that could cause an “upward bias” in the calculation of the sharp ratio. if you make the measurement interval longer, you make the annualize the standard deviation it gets bigger, which makes the sharp ratio lower. so i guess i’m confused b/c making the sharp ratio “lower” yet causing it to have an “upward bias” seems conflicting to me. but, of course, that’s because i don’t know what upward and downward bias mean…
i also remember being confused by this term with respect to NCREIF, but I can’t remember the specifics of it…
Taketwo, remember that std deviation is in the denominator of sharpe. So if you lower volatility as measured by std deviation, it will make the sharpe ratio a higher number. Anytime you get a higher number- (in this case, higher sharpe is considered better) that is an “upward bias”. With NCREIF, you have smoothed and unsmoothed versions of the index. The smoothed version uses annual appraisal data which is less frequently measured (lengthening measurement period, using more stale data) so volatility of the index is “downward biased” and if you calculate a sharpe it would be “upward biased”. Unsmoothed NCREIF uses more frequent, market based transaction datea that is more timely. So it would show greater vol most likely and not show an upward bias in Sharpe.
thanks, makes sense. the first time i read the answer in the back of the book i misread it as longer intervals having more volatility, so it confused me to say a lower sharp had an upward bias, but it was actually longer measurement interval having less volatility, resulting in a higher sharpe, thus the upward bias. got it now
“Strategies what involve taking on default risk, liquidity risk, and other forms of catastrophe risk ahve the same ability to report an upward biased sharp ratio” What is this talking about?