Variance Minimization Hedging Analysis Based on a Time-Varying Markovian DCC-GARCH Model

Abstract

Considering time-varying transition probability (TVTP), this article combines Markov regime switching with a dynamic conditional correlation generalized autoregressive conditional heteroscedasticity (DCC-GARCH) model to construct a new hedging model and study a state-dependent minimum variance hedging ratio. A two-stage maximum likelihood method is constructed to estimate the model parameters. A filtering algorithm is used in an estimation process. Empirical results on commodity futures hedging show that compared with other benchmark models, the proposed one has the best fitting effect. In addition, in terms of hedging effectiveness, the proposed model is superior to other models in most cases, which means that introducing TVTP into a DCC-GARCH model can effectively improve the performance of hedging portfolio. Note to Practitioners —This article deals with a state-dependent minimum variance hedging problem. It combines a time-varying Markov regime switching with dynamic conditional correlation generalized autoregressive conditional heteroscedasticity named DCC-GARCH to construct a new hedging model and estimates a state-dependent hedging ratio. Empirical results from commodity futures hedging show that introducing TVTP into the DCC-GARCH model can effectively reduce portfolio risk and provide better hedging performance than other traditional models, including Markov regime switching DCC-GARCH with a fixed transition probability, DCC-GARCH, ordinary least squares, naive hedging strategies, and unhedged spots. Thus, this article is of guiding significance for hedgers to fully learn the hedging rules of futures market and avoid the spots price risk.