EMD Copula based Value at Risk Estimates for Electricity Markets

In this paper we propose an Empirical Mode Decomposition Copula based approach for analyzing the portfolio risk and estimating VaR at Risk. The copula theory is introduced to analyze the time-varying microscopic dependence structure between New South Wales (NSW) and Queensland (QLD) in Australlian electricity market, and test the existence of the symmetrical dependence structure between them. The EMD algorithm is combined with copula theory to construct a new based EMD-Copula model for estimating Value-at-Risk (VaR) of electricity market. Results from the empirical analysis show that compared with the benchmark DCC-GARCH model, the proposed model outperforms the DCC-GARCH model, in terms of estimation reliability. (C) 2015 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).