一种测度金融历史事件相关性的新方法  

A New Method for Detecting the Correlation between Historical Affairs in Financial System

作　　者：郭兆纲[1] 胡振鹏[1] 

机构地区：[1]南昌大学管理学院

出　　处：《数量经济技术经济研究》2015年第9期101-118,共18页Journal of Quantitative & Technological Economics

摘　　要：边缘点分析方法是为了研究金融数据中蕴含的历史事件之间的相关性而提出的一种创新方法。该方法基于重构相空间理论,通过探测金融数据的递归图中历史数据之间的相近程度,把时间序列中点与点之间的关系转化为历史事件之间的关系,实现探测影响金融市场任一时间区域的重大历史事件的目的。本文将此方法应用于中国股票市场的分析中,得到了影响市场发展的诸如邓小平南方谈话、《证券法》的颁布与实施等重大历史事件,结果与现实情况很吻合。该方法在研究思路上属于历史回溯法,不带任何假设条件,研究过程和结果的客观性很强,也可用于其他领域的混沌时间序列分析当中。The Border Point Analysis is a creative method, which is put forward to study the correlation between the financial data and the historical affairs. Based on the reconstruction of phase space theory, the method probes the similarity of the historical data in the fi nancial data recursive graph, turns the point-to-point relation in the time series into the relation of historical affairs, and thus achieves the goals of finding the important historical af- fairs which influenced the financial market in any period. When we use this method to analyze the Chinese stock market, we get the results that fit the history and reality very well. For example, we have found the significant historical affairs which have affected the development of the market, such as the South Tour Speeches by Deng Xiaoping, the promulgation and implementation of Securities Law. The method is one of the history backtracking methods. Without any assumption, its research process and results are very objective. Therefore, the method can be used to analvze the chaos time series in other fields.

关 键 词：金融市场 递归图 类分形结构 边缘点分析 

分 类 号：F224.9[经济管理—国民经济]