This code estimates a dozen constant and time-varying copula functions for bivariate time-series (e.g. Normal, Clayton, Rotates Clayton, Plackett, Frank, Gumbel, Rotated Gumbel, Student, Symmetrised Joe-Clayton). These copulas are then compared by relying on criteria such as Log-likelihood, AIC or BIC. Besides, the code reports the plots for exceedence correlations, quantile dependence and the graphical comparison of the constant and the time-varying versions of three copulas, i.e. Normal, Gumbel and SJC. For the constant copulas, the level of tail dependence (Ldep and Udep) is also indicated.
Oxford Bulletin of Economics and Statistics (2022)
Matsuoka Hideaki