Online Program Home
My Program

Abstract Details

Activity Number: 39 - Topics in Clustering
Type: Contributed
Date/Time: Sunday, July 29, 2018 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistical Learning and Data Science
Abstract #328743 Presentation
Title: Finite Mixture-Of-Gamma Distributions: Estimation, Inference, and Model-Based Clustering
Author(s): Derek S. Young* and Xi Chen and Dilrukshi Hewage and Ricardo N. Poyanco
Companies: University of Kentucky and University of Kentucky and University of Kentucky and FONDAP Center for Genome Regulation
Keywords: cluter analysis; EM algorithm; gamma distributions; mixtools package; starting values
Abstract:

Finite mixtures of (multivariate) Gaussian distributions are one of the most commonly used distributions for model-based clustering. There is increasing recognition of mixtures of asymmetric distributions as powerful alternatives to traditional mixtures of Gaussian and mixtures of t distributions. The present work contributes to that assertion by developing a general framework for the utility of mixtures of gamma distributions. This talk presents details about maximum likelihood estimation of mixtures of gammas using an expectation-conditional-maximization (ECM). We employ the Wilson-Hilferty normal approximation as part of an effective starting value strategy for our ECM algorithm, as well as part of an effective model-based clustering strategy. We provide some simulation results as well as analyze two real data examples: an infant habituation dataset and a whole genome duplication dataset.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program