Publication Type
Journal Article
Version
publishedVersion
Publication Date
1-2006
Abstract
We consider estimation in a bivariate mixture model in which the component distributions can be decomposed into identical distributions. Previous approaches to estimation involve parametrizing the distributions. In this paper, we use a semi-parametric approach. The method is based on the exponential tilt model of Anderson (1979), where the log ratio of probability (density) functions from the bivariate components is linear in the observations. The proposed model does not require training samples, i.e., data with confirmed component membership. We show that in bivariate mixture models, parameters are identifiable. This is in contrast to previous works, where parameters are identifiable if and only if each univariate marginal model is identifiable (Teicher (1967)).
Keywords
empirical likelihood, multivariate mixture, semi-parametric, Shannon's mutual information
Discipline
Econometrics
Research Areas
Econometrics
Publication
Statistica Sinica
Volume
16
Issue
1
First Page
153
Last Page
163
ISSN
1017-0405
Publisher
Academia Sinica
Citation
LEUNG, Denis H. Y. and QIN, Jing.
Semi-parametric inference in a bivariate (multivariate) mixture model. (2006). Statistica Sinica. 16, (1), 153-163.
Available at: https://ink.library.smu.edu.sg/soe_research/435
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Additional URL
http://www3.stat.sinica.edu.tw/statistica/oldpdf/A16n19.pdf