Cross-Validation Based Covariance Shrinkage in Portfolio Selection
quantf research Working Paper Series: WP03/2014
33 Pages Posted: 2 Jun 2014
Date Written: June 1, 2014
Abstract
In this paper we investigate the applied performance of covariance shrinkage in the portfolio optimisation problem. We suggest that the optimal shrinkage coefficient should be obtained from a numerical optimisation of a function with financial interpretation. Such a function could be the minimisation of the standard deviation of the portfolio returns or the maximisation of the Sharpe Ratio of the portfolio. Empirical evidence indicates that this methodology results in portfolios with better risk/return characteristics.
Keywords: Covariance Matrix, Shrinkage Methods, Portfolio Optimisation
Suggested Citation: Suggested Citation