Estimation of a covariance matrix with zeros |
| |
Authors: | Chaudhuri Sanjay; Drton Mathias; Richardson Thomas S |
| |
Institution: | Department of Statistics and Applied Probability, Faculty of Science, 6 Science Drive 2, National University of Singapore, 117546, Singapore |
| |
Abstract: | We consider estimation of the covariance matrix of a multivariaterandom vector under the constraint that certain covariancesare zero. We first present an algorithm, which we call iterativeconditional fitting, for computing the maximum likelihood estimateof the constrained covariance matrix, under the assumption ofmultivariate normality. In contrast to previous approaches,this algorithm has guaranteed convergence properties. Droppingthe assumption of multivariate normality, we show how to estimatethe covariance matrix in an empirical likelihood approach. Theseapproaches are then compared via simulation and on an exampleof gene expression. |
| |
Keywords: | Covariance graph Empirical likelihood Graphical model Marginal independence Maximum likelihood estimation Multivariate normal distribution |
本文献已被 Oxford 等数据库收录! |
|