EGARCH-M模型在证券时间序列分析中的应用  

Application of EGARCH-M model in the stock time series analysis

作　　者：杨世康 邓世权[1] 童汝超 龙见远 吴巧霞[1] 

机构地区：[1]凯里学院,贵州凯里556011

出　　处：《无线互联科技》2016年第22期144-145,共2页Wireless Internet Technology

摘　　要：证券综合指数的对数收益率的折线图反映收益率的波动呈现出在一段时间内波动比较大,一段时间波动比较小,方差随着时间的变化而变化。在对时间序列数据进行研究的时候,通常假设随着时间的变化方差不会发生变化。但是在关注预测的精确程度时,需要了解方差的大小。文章用Eviews软件对上证综指日收盘价的对数收益率建立EGARCH-M模型,对收益率序列呈现出的波动聚集性,杠杆效应、风险与收益的关系等特征进行了分析,最终对波动率进行预测,结果表明EGARCH-M模型充分描述了波动性聚类的特点,只用很少的参数就可以把实际数据拟合得很好。该模型形式简单,容易估计,提高了对方差的预测精度,对收益率波动率建立模型对于宏观经济理论和金融理论有重要的意义。From the line chart of Logarithmic yield of stock composite index, it can be seen that the fluctuation of yields shown relatively large in a period of time and relatively small in another period, which shows that the variance changes with time changes. In the study of time serious data, it is usually assumed that the variance will not change as time changes. But when we concern about the degree of accuracy of the forecast, we need to know the size of the variance. The article use Eviews software to build EGARCH-M model for the logarithm yield of Shanghai composite index day closing price, to analyze features of volatility clustering showed in yield sequence, leverage effects, the relationship between risks and returns, and finally make a forecast to volatility. The results show that the EGARCH-M model adequately describes the characteristics of volatility clustering. With only a few parameters that can fit the actual data well. The form of the model is simple, easy to estimate, which improves the forecast accuracy of variance. It has a great importance in the building of model of yield volatility and macroeconomic theory and financial theory.

关 键 词：对数收益率 风险 聚类 EGARCH-M 杠杆效应 

分 类 号：F224[经济管理—国民经济] F832.51