Testing Whether Two Sharpe Ratios Differ Significantly
Summary
The document raises a statistical question about comparing the Sharpe ratios of two portfolios. One portfolio is described as holding similar stocks, while the other emphasizes companies majority owned by employees. The author wants to determine whether the difference between the observed ratios is statistically significant and mentions using R.
The response points to a paper on the statistics of Sharpe ratios as a resource for addressing the comparison. It does not explain a particular test, state the assumptions required, or provide calculations for the portfolios. The sample description and observed ratios are part of the question, not evidence that the difference is meaningful. A researcher would need to consult the referenced method and consider the data and dependence structure before drawing a conclusion.
Key ideas
- The question concerns statistical inference on the difference between two portfolio Sharpe ratios.
- A paper on Sharpe ratio statistics is suggested as a resource.
- The exchange gives no test procedure or result for the example portfolios.
- Inference depends on the data and assumptions used in the comparison.
Tags
Full text
# statistically compare 2 sharpe ratios # statistically compare 2 sharpe ratios I have a problem regarding comparing sharpe ratios of portfolios. As an example: I have constructed a portfolio from monthly data over 5 years, giving me a sharpe ratio of 0.85. I have a portfolio with similar stocks but with characteristics that they're majority owned by their employees, yielding sharpe ratio of 1.3. I want to test if the ratio of 1.3 is significant different than the ratio of 0.85. I'm using R for this so far. Thanks for any help beforehand Best Emil ## Answer by TickaJules (score 1) https://quant.stackexchange.com/a/70605 This paper covers the topic and should be helpful: https://alo.mit.edu/wp-content/uploads/2017/06/The-Statistics-of-Sharpe-Ratios.pdf
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.