Finding the Observed Information Matrix When Using theEMAlgorithm
Journal of the Royal Statistical Society: Series B (Statistical Methodology) · 1982 · Volume 44 · Issue 2
DOI: 10.1111/j.2517-6161.1982.tb01203.x
Abstract
A procedure is derived for extracting the observed information matrix when theEMalgorithm is used to find maximum likelihood estimates in incomplete data problems. The technique requires computation of a complete-data gradient vector or second derivative matrix, but not those associated with the incomplete data likelihood. In addition, a method useful in speeding up the convergence of theEMalgorithm is developed. Two examples are presented.
Keywords
em algorithm, observed information, maximum likelihood, speeding convergence
Enable JavaScript for citation tools and additional paper interactions.