Maximum Likelihood Estimation of the Parameters of a Multivariate Normal Distribution
Abstract
This paper provides an exposition of several altnerative techniques used to obtain maximum likelihood estimators for the parameters of a multivariate normal distribution. In particular, matrix differentiation, matrix transformations and induction are treated. These techniques are used to derive the maximum likelihood estimators of the covariances of a Wishart distribution, of the covariances when there are missing observations, and of the means under a rank constraint. Although the paper is mainly expository, some of the proofs are new.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1979
- Accession Number
- ADA073796
Entities
People
- I. Olkin
- Theodore W. Anderson
Organizations
- Stanford University