【摘要】 将小波多分辨分析与符号时间序列分析方法相结合,确定不同尺度上股指波动的主要模式与异常模式,为不同类型的投资者提供参考。通过离散小波分析得到波动序列不同尺度的细节,针对不同尺度上的细节,采用符号时间序列分析方法确定其主要模式与异常模式,并与原序列比较异同。用该方法以上证综指与深证成指两个指数的5分时序列为样本,对其"已实现"波动序列进行实证分析,以验证该方法的可行性和有效性。更多还原

【Abstract】 Symbolic time series analysis was combined with the wavelet multi-resolution analysis to the study of stock volatility.The method proposed can determine the principal and abnormal volatility change patterns of different scale components,offering reference for different types of investors.Firstly,with discrete wavelet transform,the volatility series was decomposed into different scale components.Then,the principal and abnormal volatility change patterns can be determined by symbolic time series analysis on the scale components and compared with those from original time series.The effectiveness and feasibility of the method were proved by the analysis of the volatility series of Shanghai composite stock and Shenzhen component stock which were calculated with the 5 minutes closing prices.更多还原