Robust Portfolio Selection Based on Copula Change Analysis

In this article, we construct a robust portfolio selection model based on dynamic copulas. We first use a type of dynamic copula, which contains copulas with time-varying parameters or sequence of copulas, to characterize the dynamic dependence between financial assets. Then, we use it for portfolio selection based on worst-case Conditional Value-at-Risk (WCVaR). In the empirical part we choose four representative assets from Chinese market to construct a macro asset allocation of portfolio and make the performance analysis. Results show that our method performs the best in out-of-sample tests when considering the dynamic dependence between assets and the uncertainty in the estimated model.