A general approach to heteroscedastic linear regression

DS Leslie, R Kohn, DJ Nott

Research output: Contribution to journalArticle (Academic Journal)peer-review

25 Citations (Scopus)

Abstract

Our article presents a general treatment of the linear regression model, in which the error distribution is modelled nonparametrically and the error variances may be heteroscedastic, thus eliminating the need to transform the dependent variable in many data sets. The mean and variance components of the model may be either parametric or nonparametric, with parsimony achieved through variable selection and model averaging. A Bayesian approach is used for inference with priors that are data-based so that estimation can be carried out automatically with minimal input by the user. A Dirichlet process mixture prior is used to model the error distribution nonparametrically; when there are no regressors in themodel, the method reduces to Bayesian density estimation, and we show that in this case the estimator compares favourably with a well-regarded plug-in density estimator. We also consider a method for checking the fit of the full model. The methodology is applied to a number of simulated and real examples and is shown to work well.
Translated title of the contributionA general approach to heteroscedastic linear regression
Original languageEnglish
Pages (from-to)131 - 146
Number of pages16
JournalStatistics and Computing
Volume17 (2)
DOIs
Publication statusPublished - Jun 2007

Bibliographical note

Publisher: Springer

Fingerprint Dive into the research topics of 'A general approach to heteroscedastic linear regression'. Together they form a unique fingerprint.

Cite this