Robust Portfolio Optimization Under Sampling Error: Estimation of Sampling Error in Small Eigen Value by Semidefinite Programming

9 Pages Posted: 11 Jul 2013 Last revised: 13 Sep 2013

Date Written: July 11, 2013

Abstract

The optimized portfolio that is calculated by a covariance matrix has large sensitivities to small eigen values of the covariance matrix. Estimation of sampling errors for small eigen values is quite important for fund managers who construct their portfolios from estimated covariance matrixes. If they can calculate the sampling errors, they can construct robust portfolios. In this study, we propose the method of estimation of sampling error for eigen values from error of each element of covariance matrix. And we will show how this estimation is useful to optimized portfolio.

Keywords: robust optimization, mean variance optimization, SDP, Semidefinite Programming, sampling error, eigen value, covariance matrix

JEL Classification: C61, G11

Suggested Citation

Minami, Seiji, Robust Portfolio Optimization Under Sampling Error: Estimation of Sampling Error in Small Eigen Value by Semidefinite Programming (July 11, 2013). Available at SSRN: https://ssrn.com/abstract=2292361 or http://dx.doi.org/10.2139/ssrn.2292361

Seiji Minami (Contact Author)

Resona Bank ( email )

Fukagawa Gatharia W2
5-65, Kiba 1-Chome
Tokyo, 135-8581
Japan

Register to save articles to
your library

Register

Paper statistics

Downloads
267
Abstract Views
1,566
rank
115,096
PlumX Metrics