论文部分内容阅读
预测股指时间序列突变点是在股票市场上进行投资的一个关键问题,而检测突变点是预测的基础.在检测深沪两市股指时间序列月度收益率突变点位置和个数时采用了非参数方法,该方法基于小波数据依赖门限技术.研究显示了运用Lipschitz指数解释的突变点的数学特征.使用的模型证明了小波变换模的极大值能够检测出突变点的位置,实证结果也显示出突变点的位置和个数是精确的.
It is a key problem to forecast the abrupt change of time series of stock index in the stock market, and the detection of abrupt change is the basis of prediction.When testing the location and number of abrupt change of monthly yield of Shenzhen and Shanghai stock index time series, This method is based on the wavelet data-dependent threshold technique. The research shows the mathematical characteristics of the abrupt point explained by the Lipschitz exponent. The model used proves that the maximum value of the wavelet transform mode can detect the location of the abrupt point. The empirical results also show that The location and number of mutation points are accurate.