基于MRSD Copula-GJR模型的股指期货套期保值研究

发布时间：2018-04-29 13:19

本文选题：股指期货 + 套期保值　；参考：《湖南大学》2012年硕士论文

【摘要】：最优套期保值策略一直是股指期货套期保值研究领域最为重要的课题之一，其效果主要取决于套期保值比率估计的精确程度。本文针对股指期货和股票现货收益特征及其相关特征，构建了MRSD Copula函数的GJR-Skew-t模型，用以估计股指期货与指数现货之间的最小方差套期保值比率。本文以香港、日本、新加坡、韩国和台湾5个亚洲证券市场的股指期货和股票现货作为研究对象，首先根据股指期货与股票现货收益所存在的有偏、尖峰厚尾分布以及非对称波动特征，构建GJR-skewed-t模型对股指期货与股票现货收益的条件波动率进行估计，并确定其具体边际分布。然后，利用各个市场累积分布函数值序列对MRSD Copula模型进行估计，从而得到基于MRSD Normal Copu-la-GJR模型条件下股指期货与股票现货收益的联合分布，进而计算出最优套期保值比率。最后，依据基于MRSD Normal Copula-GJR模型以及其它模型的套期保值比率，对各模型的套期保值绩效进行对比分析。研究结果表明，，动态套期保值模型的风险规避效果明显优于静态模型；根据套期保值组合方差降低百分比，基于MRSD Normal Copula-GJR模型的套期保值效果比其它动态策略有显著提升；除新加坡市场外，基于MRSD Copula函数的套期保值模型可以获得比传统模型更高的收益，这意味着该策略模型有助于降低套期保值成本。
[Abstract]:Optimal hedging strategy has always been one of the most important topics in the field of stock index futures hedging. Its effect mainly depends on the accuracy of hedge ratio estimation. In this paper, according to the characteristics of stock index futures and stock spot returns and their related characteristics, a GJR-Skew-t model of MRSD Copula function is constructed to estimate the minimum variance hedge ratio between stock index futures and index spot. This paper takes stock index futures and stock spot in five Asian stock markets of Hong Kong, Japan, Singapore, South Korea and Taiwan as research objects. Firstly, according to the bias between stock index futures and stock spot returns, The GJR-skewed-t model is constructed to estimate the conditional volatility of stock index futures and stock spot returns and to determine the specific marginal distribution. Then, the MRSD Copula model is estimated by using each market cumulative distribution function sequence, and the joint distribution of stock index futures and stock spot returns is obtained based on MRSD Normal Copu-la-GJR model, and the optimal hedging ratio is calculated. Finally, according to the hedge ratio based on MRSD Normal Copula-GJR model and other models, the hedging performance of each model is compared and analyzed. The results show that the risk avoidance effect of dynamic hedging model is obviously better than that of static model, and the hedge effect based on MRSD Normal Copula-GJR model is significantly improved compared with other dynamic strategies according to the percentage of variance reduction of hedging portfolio. In addition to the Singapore market, the hedging model based on the MRSD Copula function can achieve higher returns than the traditional model, which means that the strategy model can help to reduce the hedging cost.
【学位授予单位】：湖南大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F830.91;F224

【参考文献】