期刊
IEEE TRANSACTIONS ON SIGNAL PROCESSING
卷 69, 期 -, 页码 5681-5692出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TSP.2021.3118546
关键词
Covariance matrices; Tuning; Statistics; Sociology; Estimation; Optimization; Portfolios; Covariance matrix estimation; regularization; shrinkage; elliptical distribution; regularized discriminant analysis
资金
- Academy of Finland [298118]
The article focuses on the estimation of covariance matrices of multiple classes through the use of regularized SCM estimators. By coupling the regularization towards the pooled SCM and scaled identity matrix, the proposed techniques show promising MSE performance in scenarios where class populations follow elliptical distributions. The coupled RSCMs demonstrate comparable performance to cross-validation but with significantly faster computation time when applied on real data sets.
The estimation of covariance matrices of multiple classes with limited training data is a difficult problem. The sample covariance matrix (SCM) is known to perform poorly when the number of variables is large compared to the available number of samples. In order to reduce the mean squared error (MSE) of the SCM, regularized (shrinkage) SCM estimators are often used. In this work, we consider regularized SCM (RSCM) estimators for multiclass problems that couple together two different target matrices for regularization: the pooled (average) SCM of the classes and the scaled identity matrix. Regularization toward the pooled SCM is beneficial when the population covariances are similar, whereas regularization toward the identity matrix guarantees that the estimators are positive definite. We derive the MSE optimal tuning parameters for the estimators as well as propose a method for their estimation under the assumption that the class populations follow (unspecified) elliptical distributions with finite fourth-order moments. The MSE performance of the proposed coupled RSCMs are evaluated with simulations and in a regularized discriminant analysis (RDA) classification set-up on real data. The results based on three different real data sets indicate comparable performance to cross-validation but with a significant speed-up in computation time.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据