Header

UZH-Logo

Maintenance Infos

Optimal estimation of a large-dimensional covariance matrix under Stein’s loss


Ledoit, Olivier; Wolf, Michael (2017). Optimal estimation of a large-dimensional covariance matrix under Stein’s loss. Working paper series / Department of Economics 122, University of Zurich.

Abstract

This paper introduces a new method for deriving covariance matrix estimators that are decision-theoretically optimal within a class of nonlinear shrinkage estimators. The key is to employ large-dimensional asymptotics: the matrix dimension and the sample size go to infinity together, with their ratio converging to a finite, nonzero limit. As the main focus, we apply this method to Stein’s loss. Compared to the estimator of Stein (1975, 1986), ours has five theoretical advantages: (1) it asymptotically minimizes the loss itself, instead of an estimator of the expected loss; (2) it does not necessitate post-processing via an ad hoc algorithm (called “isotonization”) to restore the positivity or the ordering of the covariance matrix eigenvalues; (3) it does not ignore any terms in the function to be minimized; (4) it does not require normality; and (5) it is not limited to applications where the sample size exceeds the dimension. In addition to these theoretical advantages, our estimator also improves upon Stein’s estimator in terms of finite-sample performance, as evidenced via extensive Monte Carlo simulations. To further demonstrate the effectiveness of our method, we show that some previously suggested estimators of the covariance matrix and its inverse are decision-theoretically optimal in the large-dimensional asymptotic limit with respect to the Frobenius loss function.

Abstract

This paper introduces a new method for deriving covariance matrix estimators that are decision-theoretically optimal within a class of nonlinear shrinkage estimators. The key is to employ large-dimensional asymptotics: the matrix dimension and the sample size go to infinity together, with their ratio converging to a finite, nonzero limit. As the main focus, we apply this method to Stein’s loss. Compared to the estimator of Stein (1975, 1986), ours has five theoretical advantages: (1) it asymptotically minimizes the loss itself, instead of an estimator of the expected loss; (2) it does not necessitate post-processing via an ad hoc algorithm (called “isotonization”) to restore the positivity or the ordering of the covariance matrix eigenvalues; (3) it does not ignore any terms in the function to be minimized; (4) it does not require normality; and (5) it is not limited to applications where the sample size exceeds the dimension. In addition to these theoretical advantages, our estimator also improves upon Stein’s estimator in terms of finite-sample performance, as evidenced via extensive Monte Carlo simulations. To further demonstrate the effectiveness of our method, we show that some previously suggested estimators of the covariance matrix and its inverse are decision-theoretically optimal in the large-dimensional asymptotic limit with respect to the Frobenius loss function.

Statistics

Downloads

37 downloads since deposited on 14 May 2013
7 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Working Paper
Communities & Collections:03 Faculty of Economics > Department of Economics
Working Paper Series > Department of Economics
Dewey Decimal Classification:330 Economics
JEL Classification:C13
Uncontrolled Keywords:Large-dimensional asymptotics, nonlinear shrinkage estimation, random matrix theory, rotation equivariance, Stein’s loss
Language:English
Date:March 2017
Deposited On:14 May 2013 12:31
Last Modified:22 Mar 2017 16:59
Series Name:Working paper series / Department of Economics
Number of Pages:63
ISSN:1664-7041
Additional Information:Revised version
Official URL:http://www.econ.uzh.ch/static/wp/econwp122.pdf
Related URLs:http://www.econ.uzh.ch/static/workingpapers.php

Download

Download PDF  'Optimal estimation of a large-dimensional covariance matrix under Stein’s loss'.
Preview
Filetype: PDF (Revised version March 2017)
Size: 799kB