TY - JOUR
T1 - Variable selection in finite mixture of regression models with an unknown number of components
AU - Lee, Kuo Jung
AU - Feldkircher, Martin
AU - Chen, Yi Chi
N1 - Funding Information:
The authors wish to thank Yen-Chi Chen, Adrian Dobra, Bettina Grün, Darryl Holman, Chang-Jin Kim, Gian-Luca Melchiorre, Roberto León-González, Eric Zivot, and participants at the University of Washington Center for Statistics and the Social Sciences (CSSS) seminar, the 2019 North American Meeting of the Econometric Society, the annual meeting of Taiwan Econometric Society for helpful comments. The authors are also very grateful to the Editor, anonymous Associate Editor and referees for their helpful comments that have substantially improved this article. This research was supported by the Ministry of Science and Technology of the Republic of China (MOST 108-2118-M-006-002 for Lee and MOST 107-2410-H-006-014 for Chen).
Funding Information:
The authors wish to thank Yen-Chi Chen, Adrian Dobra, Bettina Grün, Darryl Holman, Chang-Jin Kim, Gian-Luca Melchiorre, Roberto León-González, Eric Zivot, and participants at the University of Washington Center for Statistics and the Social Sciences (CSSS) seminar, the 2019 North American Meeting of the Econometric Society, the annual meeting of Taiwan Econometric Society for helpful comments. The authors are also very grateful to the Editor, anonymous Associate Editor and referees for their helpful comments that have substantially improved this article. This research was supported by the Ministry of Science and Technology of the Republic of China ( MOST 108-2118-M-006-002 for Lee and MOST 107-2410-H-006-014 for Chen).
Publisher Copyright:
© 2021 Elsevier B.V.
PY - 2021/6
Y1 - 2021/6
N2 - A Bayesian framework for finite mixture models to deal with model selection and the selection of the number of mixture components simultaneously is presented. For that purpose, a feasible reversible jump Markov Chain Monte Carlo algorithm is proposed to model each component as a sparse regression model. This approach is made robust to outliers by using a prior that induces heavy tails and works well under multicollinearity and with high-dimensional data. Finally, the framework is applied to cross-sectional data investigating early warning indicators. The results reveal two distinct country groups for which estimated effects of vulnerability indicators vary considerably.
AB - A Bayesian framework for finite mixture models to deal with model selection and the selection of the number of mixture components simultaneously is presented. For that purpose, a feasible reversible jump Markov Chain Monte Carlo algorithm is proposed to model each component as a sparse regression model. This approach is made robust to outliers by using a prior that induces heavy tails and works well under multicollinearity and with high-dimensional data. Finally, the framework is applied to cross-sectional data investigating early warning indicators. The results reveal two distinct country groups for which estimated effects of vulnerability indicators vary considerably.
UR - http://www.scopus.com/inward/record.url?scp=85100251906&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85100251906&partnerID=8YFLogxK
U2 - 10.1016/j.csda.2021.107180
DO - 10.1016/j.csda.2021.107180
M3 - Article
AN - SCOPUS:85100251906
VL - 158
JO - Computational Statistics and Data Analysis
JF - Computational Statistics and Data Analysis
SN - 0167-9473
M1 - 107180
ER -