47 Pages Posted: 4 Aug 2017
Date Written: August 3, 2017
We show how to conduct asymptotically valid tests of model comparison when the extent of model mispricing is gauged by the squared Sharpe ratio improvement measure. This is equivalent to ranking models on their squared Sharpe ratios. Mimicking portfolios can be substituted for any non-traded model factors and estimation error in the portfolio weights is taken into account in the statistical inference. A parsimonious four-factor model that supplements the market with cash profitability and momentum factors, along with a monthly-updated version of the usual value spread, emerges as the dominant model over the period 1972-2015.
JEL Classification: G10, G12
Suggested Citation: Suggested Citation
Barillas, Francisco and Kan, Raymond and Robotti, Cesare and Shanken, Jay A., Model Comparison with Sharpe Ratios (August 3, 2017). Rotman School of Management Working Paper No. 3013149. Available at SSRN: https://ssrn.com/abstract=3013149