Robust mixture regression model fitting by Laplace distribution

Date

2013-09-27

Journal Title

Journal ISSN

Volume Title

Publisher

Kansas State University

Abstract

A robust estimation procedure for mixture linear regression models is proposed in this report by assuming the error terms follow a Laplace distribution. EM algorithm is imple- mented to conduct the estimation procedure of missing information based on the fact that the Laplace distribution is a scale mixture of normal and a latent distribution. Finite sample performance of the proposed algorithm is evaluated by some extensive simulation studies, together with the comparisons made with other existing procedures in this literature. A sensitivity study is also conducted based on a real data example to illustrate the application of the proposed method.

Description

Keywords

EM algorithm, Laplace distribution, Least absolute deviation, Mixture regression model

Graduation Month

December

Degree

Master of Science

Department

Department of Statistics

Major Professor

Weixing Song

Date

2013

Type

Report

Citation