基于小波多分辨分析的协整建模理论与方法的扩展;

为讨论经济及金融变量的多尺度行为，描述变量之间在不同时间尺度上的长期均衡关系，将小波多分辨分析引入协整建模理论，提出多分辨协整和多分辨误差校正模型两个概念，给出相应建模方法，克服了传统的协整建模理论无法揭示蕴含在变量内部的多时间尺度信息的缺陷。多分辨协整建模能够更加细致地捕获经济或金融变量在不同时间尺度上的关系，对两大股指的实证研究也支持了这一点。

Expansion of Modeling Theory and Method for Cointegration Based on Wavelet Multiresolution Analysis

In order to discuss the multiresolution character of economic or financial variable and capture the long run equilibrium relationship in different time scale,multiresolution analysis of wavelet has been introduced into the theory of cointegration modeling.The new conceptions of multiresolution cointegrtion and multiresolution error correction model are put forward in the paper.The method of multiresolution cointegration modeling which overcomes the shortcomings of traditional cointegration theory in discovering the information of multi-time scale contained in the variables is dicussed in detail.The new methods can capture the relationship between variables in different time scales,which has been proved by the empirical study.