Forums  > Risk Management  > Omega Ratio  
Page 1 of 1
Display using:  


Total Posts: 448
Joined: May 2012
Posted: 2015-11-04 18:35

has any of you found any real advantage in using Omega Ratios vs. Sharpe?

I know the theory. I am just asking if any of you did find it useful in real life, including when you had an analyst or a prospect looking at your numbers.

The only thing that counts: Can you make money?


Total Posts: 491
Joined: Mar 2005
Posted: 2016-09-26 00:27
I have used it in the past (more Sortino actually) for a specific, niche portfolio proposition. At the end of the day they seemed to go for the team over the risk picture and parameters. Chug Beer

All the soul of man is resolution, which in valiant men falters never, until their last breath.


Total Posts: 448
Joined: May 2012
Posted: 2016-09-26 10:37
Thanks, I was looking for alternatives to the constraints of the Sharpe Ratio, however FWIW we may have to live with Sharpe in the end... not many care about Omega Ratio, it appears...

The only thing that counts: can you make money?


Total Posts: 1074
Joined: May 2004
Posted: 2016-09-26 11:36
Keep it simple and stick to Sharpe, unless you are doing something extremely non linear (OTM options, etc...) for which Sharpe is obviously misleading.

We all know there is some amount of skewness/kurtosis in the returns of any strategy, and learn to take Sharpe ratios with a grain of salt. Just like Black-Scholes + experience is better than fancy models in many cases, Sharpe + experience is better than fancy ratios. Know your weapon.

"Earth: some bacteria and basic life forms, no sign of intelligent life" (Message from a type III civilization probe sent to the solar system circa 2016)


Total Posts: 448
Joined: May 2012
Posted: 2016-09-26 11:42
I agree, NeroTulip.

Just had in my mind a meeting with a very successful manager, years ago, and his fixation that a Sharpe above 2 could turn a small amount of money into a fortune, so the only way for him to measure a strategy was the Sharpe Ratio. He would not care about anything else.

The only thing that counts: can you make money?


Total Posts: 93
Joined: Apr 2005
Posted: 2017-06-12 12:30
I have found both empirical Omega and Sortino subject to over-fitting if your data set allows for extremely low or no downside in the in sample period - a distributional / copula approach seems to work better especially out of sample

see Good Bad , Ugly of Risk Stats and SFA Score as a RAPM
Previous Thread :: Next Thread 
Page 1 of 1