论文部分内容阅读
本文提出了样本内和样本外密度预测评估的数据驱动平滑检验(data-driven smooth test)方法,并分别采用Newey-Tauchen的方法以及West-McCracken的方法来纠正参数估计对样本内和样本外密度预测评估的影响。运用本文提出的检验方法,我们比较了各种最大熵GARCH模型对中国三个股指数据(香港恒生指数、上证综合指数和台湾加权指数)的样本内和样本外预测绩效。结果显示:(1)最大熵GARCH模型可以用来刻画中国股指数据的典型化事实,GARCH模型中考虑了厚尾和偏态特征的Pearson IV分布对中国股指收益率的样本外预测绩效是很重要的;(2)具有较好样本内拟合优度和样本内预测效果的模型未必有很好的样本外密度预测效果,考虑到样本外预测的重要性,实际应用中我们应采用具有较好样本外预测效果的模型。
In this paper, a data-driven smooth test method for predicting the out-of-sample and out-of-sample density is proposed. The Newey-Tauchen method and the West-McCracken method are respectively used to correct the influence of parameter estimation on in-sample and out- Impact of forecasting. Using the test method proposed in this paper, we compare the intra-and extra-sample predictive performance of various maximum entropy GARCH models for three Chinese stock index data (Hong Kong Hang Seng Index, Shanghai Composite Index and Taiwan Weighted Index). The results show that: (1) The maximum entropy GARCH model can be used to characterize the typical stock index data in China. The Pearson IV distribution in the GARCH model, which takes into account the characteristics of thick tail and skewness, is important for the out-of-sample forecast performance of China’s stock index ; (2) The model with good in-sample goodness of fit and in-sample prediction effect may not have a good prediction effect of out-sample density. Considering the importance of out-of-sample prediction, we should adopt a better Model of Out-of-sample Prediction.