Bayesian inference in a heteroscedastic replicated measurement error model using heavy-tailed distributions

• Published in: Journal of statistical computation and simulation Vol. 87; no. 15; pp. 2915 - 2928
• Main Authors: Cao, Chunzheng, Chen, Mengqian, Zhu, Xiaoxin, Jin, Shaobo
• Format: Journal Article
• Language: English
• Published: Abingdon Taylor & Francis 13.10.2017; Taylor & Francis Ltd
• Subjects: Bayesian analysis; Computer simulation; Error analysis; Heteroscedasticity; Markov Chain Monte Carlo; Markov chains; Monte Carlo simulation; multiple regression; Normal distribution; Outliers (statistics); Parameter estimation; Regression analysis; robustness; scale mixtures of normal; Statistical inference
• ISSN: 0094-9655; 1563-5163; 1563-5163
• DOI: 10.1080/00949655.2017.1349131

We introduce a multivariate heteroscedastic measurement error model for replications under scale mixtures of normal distribution. The model can provide a robust analysis and can be viewed as a generalization of multiple linear regression from both model structure and distribution assumption. An efficient method based on Markov Chain Monte Carlo is developed for parameter estimation. The deviance information criterion and the conditional predictive ordinates are used as model selection criteria. Simulation studies show robust inference behaviours of the model against both misspecification of distributions and outliers. We work out an illustrative example with a real data set on measurements of plant root decomposition.