Efficient estimation of a semiparametric dynamic copula model

Outline Introduction Semi-parametric dynamic copula Motivation Local likelihood estimation Variance of the estimator Bias of the estimator Bandwidth selection Estimation of joint likelihood Modeling of marginal distributions Simulations and applications Simulations Empirical example Conclusions Problems and Solutions Problems Modeling dependence is critical for financial time series Model volatility of returns of financial assets Different approaches are used to model correlations, e.g. Copulas Allow to model nonlinear dependence Look beyond correlation (i.e., linear dependence), which is required for non-elliptical distributions Problems and Solutions Problems Modeling dependence is critical for financial time series Model volatility of returns of financial assets Different approaches are used to model correlations, e.g. Copulas Allow to model nonlinear dependence Look beyond correlation (i.e., linear dependence), which is required for non-elliptical distributions Problems and Solutions Problems Modeling dependence is critical for financial time series Model volatility of returns of financial assets Different approaches are used to model correlations, e.g. Copulas Allow to model nonlinear dependence Look beyond correlation (i.e., linear dependence), which is required for non-elliptical distributions Problems and Solutions Problems Modeling dependence is critical for financial time series Model volatility of returns of financial assets Different approaches are used to model correlations, e.g. Copulas Allow to model nonlinear dependence Look beyond correlation (i.e., linear dependence), which is required for non-elliptical distributions Problems and Solutions Problems Modeling dependence is critical for financial time series Model volatility of returns of financial assets Different approaches are used to model correlations, e.g. Copulas Allow to model nonlinear dependence Look beyond correlation (i.e., linear dependence), which is required for non-elliptical distributions Problems and Solutions Problems Modeling dependence is critical for financial time series Model volatility of returns of financial assets Different approaches are used to model correlations, e.g. Copulas Allow to model nonlinear dependence Look beyond correlation (i.e., linear dependence), which is required for non-elliptical distributions Problems and Solutions Problems Modeling dependence is critical for financial time series Model volatility of returns of financial assets Different approaches are used to model correlations, e.g. Copulas Allow to model nonlinear dependence Look beyond correlation (i.e., linear dependence), which is required for non-elliptical distributions Problems and Solutions Problems Modeling dependence is critical for financial time series Model volatility of returns of financial assets Different approaches are used to model correlations, e.g. Copulas Allow to model nonlinear dependence Look beyond correlation (i.e., linear dependence), which is required for non-elliptical distributions Outline Introduction Semi-parametric dynamic copula Motivation …