Multi Anchor Point Shrinkage for the Sample Covariance Matrix (Extended Version)

TL;DR

This paper develops a generalized shrinkage framework for estimating the covariance matrix in high-dimensional, limited sample size settings, improving upon existing methods like the GPS estimator by incorporating additional information.

Contribution

It introduces a flexible shrinkage approach that utilizes multiple anchor points and prior information to enhance covariance matrix estimation in high-dimensional portfolios.

Findings

The new framework outperforms the GPS estimator in numerical experiments.

Incorporating sector and prior information improves estimation accuracy.

Theoretical results support the effectiveness of the generalized shrinkage approach.

Abstract

Portfolio managers faced with limited sample sizes must use factor models to estimate the covariance matrix of a high-dimensional returns vector. For the simplest one-factor market model, success rests on the quality of the estimated leading eigenvector "beta". When only the returns themselves are observed, the practitioner has available the "PCA" estimate equal to the leading eigenvector of the sample covariance matrix. This estimator performs poorly in various ways. To address this problem in the high-dimension, limited sample size asymptotic regime and in the context of estimating the minimum variance portfolio, Goldberg, Papanicolau, and Shkolnik developed a shrinkage method (the "GPS estimator") that improves the PCA estimator of beta by shrinking it toward a constant target unit vector. In this paper we continue their work to develop a more general framework of shrinkage targets that allows the practitioner to make use of further information to improve the estimator. Examples include sector separation of stock betas, and recent information from prior estimates. We prove some precise statements and illustrate the resulting improvements over the GPS estimator with some numerical experiments.