We study the sample estimation risk of the traditional Sharpe ratio without the restrictive assumption of normality for return series. We derive analytical results for the approximate bias and variance of the sample Sharpe ratio in terms of the underlying distribution parameters. The results clarify several misinterpretations existing in the literature. A Monte Carlo study shows that our bias and variance formulae approximate the true moments of the sample Sharpe ratio remarkably well. We propose using the analytical results to design an estimation risk-adjusted Sharpe ratio. An empirical study of mutual fund performance shows that using the adjusted Sharpe ratio gives a quite different performance ranking of those traditionally top-ranked funds.
Description
Estimation Risk-Adjusted Sharpe Ratio and Fund Performance Ranking Under a General Return Distribution by Yong Bao :: SSRN
%0 Journal Article
%1 baoestimation
%A Bao, Yong
%D 2009
%J Journal of Financial Econometrics
%K quantfinance sharpe statistics
%N 2
%P 152-173
%R 10.1093/jjfinec/nbn022
%T Estimation Risk-Adjusted Sharpe Ratio and Fund Performance Ranking Under a General Return Distribution
%U https://doi.org/10.1093/jjfinec/nbn022
%V 7
%X We study the sample estimation risk of the traditional Sharpe ratio without the restrictive assumption of normality for return series. We derive analytical results for the approximate bias and variance of the sample Sharpe ratio in terms of the underlying distribution parameters. The results clarify several misinterpretations existing in the literature. A Monte Carlo study shows that our bias and variance formulae approximate the true moments of the sample Sharpe ratio remarkably well. We propose using the analytical results to design an estimation risk-adjusted Sharpe ratio. An empirical study of mutual fund performance shows that using the adjusted Sharpe ratio gives a quite different performance ranking of those traditionally top-ranked funds.
@article{baoestimation,
abstract = {We study the sample estimation risk of the traditional Sharpe ratio without the restrictive assumption of normality for return series. We derive analytical results for the approximate bias and variance of the sample Sharpe ratio in terms of the underlying distribution parameters. The results clarify several misinterpretations existing in the literature. A Monte Carlo study shows that our bias and variance formulae approximate the true moments of the sample Sharpe ratio remarkably well. We propose using the analytical results to design an estimation risk-adjusted Sharpe ratio. An empirical study of mutual fund performance shows that using the adjusted Sharpe ratio gives a quite different performance ranking of those traditionally top-ranked funds.},
added-at = {2018-05-23T05:38:01.000+0200},
author = {Bao, Yong},
biburl = {https://www.bibsonomy.org/bibtex/2193483eb793367c00828d12311eb995a/shabbychef},
description = {Estimation Risk-Adjusted Sharpe Ratio and Fund Performance Ranking Under a General Return Distribution by Yong Bao :: SSRN},
doi = {10.1093/jjfinec/nbn022},
interhash = {c9a1e767ff13facb9024f457ad7b76e4},
intrahash = {193483eb793367c00828d12311eb995a},
journal = {Journal of Financial Econometrics},
keywords = {quantfinance sharpe statistics},
language = {English},
location = {https://ssrn.com/paper=1365736},
number = 2,
pages = {152-173},
timestamp = {2018-05-23T05:39:43.000+0200},
title = {Estimation Risk-Adjusted {S}harpe Ratio and Fund Performance Ranking Under a General Return Distribution},
url = {https://doi.org/10.1093/jjfinec/nbn022},
volume = 7,
year = 2009
}