Maximum Likelihood Estimation Via the ECM Algorithm: Computing the Asymptotic Variance

David A. van Dyk and Xiao-Li Meng
Department of Statistics, The University of Chicago, Chicago, IL 60637, U.S.A.
Donald B. Rubin
Department of Statistics, Harvard University, Cambridge, MA 02138, U.S.A.

This paper provides detailed theory, algorithms, and illustrations for computing asymptotic variance-covariance matrices for maximum likelihood estimates using the ECM algorithm (Meng and Rubin, 1993). This Supplemented ECM (SECM) algorithm is developed as an extension of the Supplemented EM (SEM) algorithm (Meng and Rubin, 1991a). Explicit examples are given, including one that demonstrates SECM, like SEM, has a powerful internal error detecting system for the implementation of the parent ECM or in SECM itself.

Key words: Contingency Table; Convergence Rate; EM Algorithm; Fisher Information; Incomplete Data; IPF; Missing Data; SEM Algorithm

Return to David van Dyk's homepage.