The Copula-GARCH model of conditional dependencies: An international stock market application

摘要: Abstract Modeling the dependency between stock market returns is a difficult task when follow complicated dynamics. When are non-normal, it often simply impossible to specify multivariate distribution relating two or more return series. In this context, we propose new methodology based on copula functions, which consists in estimating first univariate distributions and then joining distribution. such parameter can easily be rendered conditional time varying. We apply daily of four major markets. Our results suggest that depends past realizations for European pairs only. For these markets, found widely affected move same direction than they opposite directions. dynamics also suggests higher persistent