Mixture models in measurement error problems, with reference to epidemiological studies (1994)
| Venue: | Institute of International Economics Project on International Competition Policy," COM/DAFFE/CLP/TD(94)42 |
| Citations: | 5 - 2 self |
BibTeX
@INPROCEEDINGS{Richardson94mixturemodels,
author = {Sylvia Richardson and Laurent Leblond and Isabelle Jaussent and Peter J. Green},
title = {Mixture models in measurement error problems, with reference to epidemiological studies},
booktitle = {Institute of International Economics Project on International Competition Policy," COM/DAFFE/CLP/TD(94)42},
year = {1994},
publisher = {}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
This paper focuses on Bayesian measurement error problems and on the question of the specication of the prior distribution of the unknown covariates. It presents a exible semiparametric model for this distribution based on a mixture of normals with an unknown number of components. Implementation of this prior model as part of a full Bayesian analysis of measurement error problems is described in classical set-ups encountered in epidemiological studies: logistic regression between unknown covariates and outcome, with normal or lognormal error model and a validation group. The feasibility of this combined model is tested and its performance is demonstrated in a simulation study that includes an assessment of the inuence of misspecication of the prior distribution of the unknown covariates and a comparison with the semiparametric ML method of Roeder, Carroll and Lindsay (1996). Finally, the methodology is illustrated on a data set on coronary heart disease and blood level of cholester
