An Effective Strategy for Initializing the EM Algorithm in Finite Mixture Models

• Activity Number: 442
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistical Consulting
• Author(s): Semhar Michael* and Volodymyr Melnykov
• Companies: The University of Alabama and The University of Alabama
• Keywords: finite mixture models ; EM algorithm ; initialization ; model averaging

Abstract

Finite mixture models represent one of the most popular tools for modeling heterogeneous data. The traditional approach for parameter estimation is based on maximizing the likelihood function. The direct optimization is often troublesome due to the complex likelihood structure. The expectation maximization algorithm proves to be an effective remedy that allows alleviating this issue. The solution obtained by this procedure is entirely driven by the choice of starting parameter values. This dictates the importance of an effective initialization strategy. Despite efforts undertaken in this area, there is no uniform winner found and practitioners tend to ignore the issue, often finding misleading or erroneous results. In this paper, we propose a simple yet effective tool for initializing the expectation maximization algorithm in the mixture modeling setting. The idea is based on model averaging and proves to be efficient in detecting correct solutions even in those cases when competitors demonstrate degrading performance. The utility of the proposed methodology is shown through comprehensive simulation studies.