Asymptotic Properties of a Stochastic EM Algorithm for Estimating Mixing Proportions
Abstract
The purpose of this paper is to study the asymptotic behavior of the Stochastic EM algorithm (SEM) in a simple particular case within the mixture context. We consider the estimation of the mixing proportion p of a two- component mixture of densities assumed to be known. We establish that the stationary distribution of the ergodic Markov chain generated by SEM is asymptotic, as the sample size N tends to infinity, to a Gaussian distribution with mean the consistent maximum likelihood estimate of p and variance proportional to N-1/2. Similarly, we determine the limiting distributions of two sequential versions of SEM and study their asymptotic relative efficiency.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1991
- Accession Number
- ADA246263
Entities
People
- Gilles Celeux
- Jean Diebolt
Organizations
- University of Washington