作　　者： ;

机构地区： 华南理工大学

出　　处： 《管理科学学报》 2012年第5期 19-31,共13页

摘　　要： 基于幂转换以及不设定扰动项的具体相关结构和分布形式,构建了半参数的短期预测模型来预测中国股市的波动率.模型采用基于极值估计量的两阶段估计法进行估计,估计方法的小样本性质表现良好.此外,还通过具有Bootstrap特性的SPA检验实证比较了新模型与其他6种预测模型的预测精度.实证结果表明,在各种损失函数下,半参数短期预测模型是预测中国股市波动率精度最高的模型. A A semiparametric time series model through power transformation and not setting the dependency structure and distributional form of its error component is proposed to forecast the volatility in Chinese stock markets. The model is estimated by an extreme value estimator based the two-stage estimation method and this estimation method works fairely well in finite samples. In addition, a bootstrap SPA test is used to evaluate the predicting accuracy for the proposed model and other 6 models. The empirical results show that, under various loss functions, the proposed model is the best model for volatility forecasts among the 6 models in Chinese stock markets.

关 键 词： 已实现波动率 半参数模型 SPA检验

分 类 号： [F830]

领　　域： []