随机波动模型估计及在金融风险防范中的应用

Estimation of stochastic Volatility Model and lts Application for Avoiding Financial Risk

对随机波动 (SV)模型提出了一种基于禁忌遗传算法的伪极大似然 (TSGA- QML )估计 .Monte Carlo试验表明这种方法在参数估计与波动估计上都有较好效果 .利用这一方法对上海股市收益进行了波动分析 ,发现上海股市的收益具有很高的波动持续性 .

This paper presents a quasi maximum likelihood estimation of stochastic volatility model based on tabu search genetic algorithms, and monte carlo experiments show that the method performs well with respect to both parameter estimates and volatility estimates We also illustrate the method by analyzing daily stock return on the shanghai stock exchange and find the high persistence in volatility for the return series