Estimation of a covariance matrix with zeros

• 成果类型: Article
• 署名作者: Chaudhuri, Sanjay; Drton, Mathias; Richardson, Thomas S.
• 署名单位: National University of Singapore; University of Chicago; University of Washington; University of Washington Seattle
• 刊物名称: BIOMETRIKA
• ISSN/ISSBN: 0006-3444
• DOI: 10.1093/biomet/asm007
• 发表日期: 2007
• 页码: 199216
• 关键词: seemingly unrelated regressions; maximum-likelihood-estimation; distributions; models; expression; EQUATIONS; graphs

摘要

We consider estimation of the covariance matrix of a multivariate random vector under the constraint that certain covariances are zero. We first present an algorithm, which we call iterative conditional fitting, for computing the maximum likelihood estimate of the constrained covariance matrix, under the assumption of multivariate normality. In contrast to previous approaches, this algorithm has guaranteed convergence properties. Dropping the assumption of multivariate normality, we show how to estimate the covariance matrix in an empirical likelihood approach. These approaches are then compared via simulation and on an example of gene expression.