基于时变Markov的DCC-GARCH模型最小风险套期保值研究

考虑Markov状态转移概率的时变特征，在传统DCC-GARCH基础上，提出基于Markov时变转移概率的DCC-GARCH模型（TVTP-DCC-GARCH）研究最小风险套期保值比例的估计方法，并利用两阶段极大似然法对模型参数进行估计。进一步分别从样本内和样本外估计沪深300指数期货和现货的最优套期保值比率，对套期保值的绩效进行检验，并将检验结果分别与Markov转移概率恒定的DCC-GARCH（FTP-DCC-GARCH）、DCC-GARCH、OLS、1：1完全套期保值以及无套期保值的沪深300指数现货的绩效进行对比。实证结果表明，利用基于Markov状态转移的DCC-GARCH模型研究沪深300指数期货的套期保值问题具有一定合理性，且在参数估计中TVTP-DCC-GARCH模型的拟合效果最佳；在套期保值有效性方面，TVTP-DCC-GARCH模型优于其他模型，说明在DCC-GARCH模型中引入时变状态转移概率能够有效提高套期保值组合的绩效。

Research on Variance Minimization Hedging Based on Time-Varying Markov DCC-GARCH Model

Stock market fluctuate frequently because of information shocks caused by sudden events resulting from economic, political, or natural disasters. Therefore, hedging the stock market has always been the popular research topic. Futures market, as an important part of the financial markets, is always used to operate hedging strategies in order to realize the risk transfer. The key problem of hedging theory is to determine the optimal hedging ratio. In this study, using a regime switching framework, a new estimation method of minimum risk hedging ratio is proposed. Then, taking the actual data of CSI300 index futures and spot as samples, the hedging ratios are estimated respectively from both in sample and out of sample. Compared with traditional hedging methods, the hedging performance of this new method is tested. This study is of great significance for hedgers to fully understand the hedging rules of futures market and avoid the volatility risk of spot price effectively.In the first part, considering the time varying characteristic of Markov regime transition probability, based on the traditional DCC-GARCH, a Markov regime switching DCC-GARCH model with time varying transition probability (TVTP-DCC-GARCH) is presented to study on the estimation method of minimum variance hedge ratio. Two-stage maximum likelihood method is used to estimate the parameters of the model. In the second part, with the actual data of CSI300 index futures and spot in sample, the hedging ratios of TVTP-DCC-GARCH is estimated, and the hedging performance is compared with other models, including a MRS-DCC-GARCH with a fixed transition probability (FTP-DCC-GARCH), DCC-GARCH, OLS, naïve hedging strategy and indices spot with no hedging. Furthermore, one-step-ahead forecasts out of sample are produced to forecast the hedging ratios of TVTP-DCC-GARCH and the hedging performance of the above models is checked.In summary, the DCC-GARCH model based on Markov regime switching is reasonable to study the hedging problem of CSI300 index future, and the TVTP-DCC-GARCH model has the best fitting effect. Thus, it is necessary to build a hedging model based on Markov with time varying transition probability, and explore the impact of time varying transition probability on the optimal hedging ratio in the futures market. In addition, in terms of hedging effectiveness, TVTP-DCC-GARCH model is superior to other models, which means that introducing the time varying transition probability into DCC-GARCH model can effectively improve the performance of hedging portfolio.