|
Activity Number:
|
414
- Model Building and Selection
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 1, 2017 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
|
Abstract #322897
|
|
|
Title:
|
Gaussian Mixture Modeling Under Measurement Uncertainty
|
|
Author(s):
|
Shuchismita Sarkar* and Volodymyr Melnykov and Rong Zheng
|
|
Companies:
|
University of Alabama and The University of Alabama and University of Alabama
|
|
Keywords:
|
Finite mixture modeling ;
Model based clustering ;
Gaussian mixture model ;
EM algorithm
|
|
Abstract:
|
Finite mixtures are popular in statistics due to their flexibility in modeling heterogeneous data. Model-based clustering assumes that there is a one-to-one association between mixture components and data groups also known as clusters. The situation where observations are not known with certainty is considered. We propose a model capable of taking into account such uncertainty. The developed methodology is illustrated on Gaussian mixture models. We demonstrate how parameter estimates can be obtained based on the traditional expectation-maximization algorithm. The approach is illustrated on several artificial and real-life data sets.
|
Authors who are presenting talks have a * after their name.
