Bayesian mixture copula estimation and selection

Mixture copula is a popular and essential tool for studying complex dependencies among variables. However, selecting the correct mixture models often involve repeated testing and estimations, which could cause effort and time. In this paper, we propose a method that would enable us to select and estimate the correct mixture copula simultaneously. This is done by first overfitting the model, then conducting the Bayesian estimation. We verify the correctness of our approach by numerical simulation. Finally, the real data analysis is done by studying the dependencies among three major financial markets. MSC 2020: 6208, 62F15, 62H05, 62H30