Intl. Conference on Economics, Finance & Business, Paris

APPLICATION OF THE BAYESIAN COPULA-GARCH MODEL WITH DYNAMIC GUMBEL COPULA TO MODEL THE RATE OF RETURN ON THE FINANCIAL MARKET

JUSTYNA MOKRZYCKA

Abstract:

Modeling the volatility and dependence structure of financial time series is a significant area of interest for both researchers and practitioners in financial markets. In this study, we introduce a formal Bayesian statistical model: the Copula-GARCH with dynamic Gumbel copula (an asymmetric copula), and conduct a comparative analysis with various MGARCH structures and other Copula-GARCH models. Our work includes formal Bayesian model comparison, employing the computation of marginal data density values as overarching indicators of Bayesian model adequacy. The dataset comprises bivariate logarithmic daily rates of return stock market sub-indices.

Keywords: Bayesian Copula-GARCH model