Home
Resource
Essays on Bayesian inference in regression models with non-normal error terms
PDF
PDF format is widely accepted and good for printing.
Plug-in required
PDF-1(414.19 kb)
Citation & Export
View Usage Statistics
Staff View

Citation & Export
Hide

Simple citation

Shimizu, Haruhiko. Essays on Bayesian inference in regression models with non-normal error terms. Retrieved from https://doi.org/doi:10.7282/T3NG4PPK

Export

Click here for information about Citation Management Tools at Rutgers.

Statistics
Hide

Description

TitleEssays on Bayesian inference in regression models with non-normal error terms
NameShimizu, Haruhiko (author); Tsurumi, Hiroki (chair); Hughes, Joseph P (internal member); Landon-Lane, John (internal member); Goldman, Elena (outside member); Rutgers University; Graduate School - New Brunswick
Date Created2011
Other Date2011-10 (degree)
SubjectEconomics, Regression analysis, Bayesian statistical decision theory
Extentix, 37 p. : ill.
DescriptionThe dissertation consists of three essays on regression models with non-normal error terms. In the first essay, I consider the Bayesian analysis of regression model with the error term following the exponential power distribution (EPD) that includes normal and Laplace distributions as special cases. Many empirical data are known to follow non-normal distributions. If we use a distribution that is flexible enough to include the normal distribution as a special case, we may be able to explain the data better. In estimating the regression model I use a modified efficient jump algorithm for Markov Chain Monte Carlo (MCMC) draws of the parameters of the epd distribution. I also discuss the unbiasedness of the least absolute deviation (LAD) estimator. The LAD estimator of the regression coefficients turns out to be unbiased except for the constant term even if the error term has a skewed distribution. Second essay is about the censored model known as the Tobit. Yu and Stander (2007) proposed a Bayesian quantile regression model using the asymmetric Laplace distribution (ALD). Yu and Staner (2007) formulated the likelihood function based on the quantiles. I construct the likelihood function based on the usual way the Tobit regression model is defined in econometrics. I show by using Monte Carlo simulation experiments that the way I constructed the likelihood function works better than Yu and Stander's likelihood function judged by the mean squared errors. I present an application using the data on hours worked by women. In the third essay, I evaluate the performance of a popular Bayesian model selection criteria known as the deviance information criterion (DIC) proposed by Spiegelhalter et.al (2002). I use the Tobit regression models. I find from the simulation analysis that if the data are generated from the ALD, the DIC consistently chooses the correct model. If the data is generated from the normal distribution, the performance of the DIC depends on the signs of the regression coefficients. As a real data analysis the Tobit regression model is applied to the number of medals obtained in the Olympic Games by participating countries.
NotePh.D.
NoteIncludes bibliographical references
NoteIncludes vita
Noteby Haruhiko Shimizu
Genretheses, ETD doctoral
Persistent URLhttps://doi.org/doi:10.7282/T3NG4PPK
Languageeng
CollectionGraduate School - New Brunswick Electronic Theses and Dissertations
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work.
Version 8.5.5
Rutgers University Libraries - Copyright ©2024