Please use this identifier to cite or link to this item: http://cmuir.cmu.ac.th/jspui/handle/6653943832/65524
Title: Bayesian approach for mixture copula model
Authors: Sukrit Thongkairat
Woraphon Yamaka
Songsak Sriboonchitta
Keywords: Computer Science
Issue Date: 1-Jan-2019
Abstract: © Springer Nature Switzerland AG 2019. This paper aims to use the Bayesian estimation as an alternative method for formulating and estimating mixed copula models. This method has claimed to be more efficient than the conventional maximum likelihood estimator as it can deal with the high dimension copula and large parameter estimates under limited sample. In this study, we present various mixed copula functions constructed from both Elliptical and Archimedean copulas. We employ a simulation study to investigate the performance of this estimator for comparison with the maximum likelihood estimator. The results show that the Bayesian estimation is considerably more accurate than maximum likelihood estimator in various scenarios. Finally, we extend the Bayesian mixed copula to the real data and show that our approach perform well in this real data analysis.
URI: https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85065623405&origin=inward
http://cmuir.cmu.ac.th/jspui/handle/6653943832/65524
ISSN: 1860949X
Appears in Collections:CMUL: Journal Articles

Files in This Item:
There are no files associated with this item.


Items in CMUIR are protected by copyright, with all rights reserved, unless otherwise indicated.