运用变点理论对连涨连跌收益率的Bayes分析  

Application of change point theory in successive rises and falls of returns with Bayes analysis

作　　者：黄飞[1] 谭常春[1] 

机构地区：[1]合肥工业大学数学学院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2014年第2期248-252,共5页Journal of Hefei University of Technology：Natural Science

基　　金：安徽省自然科学基金资助项目(1208085QA12);全国统计科研计划重点资助项目(2012LZ009);安徽省高等学校省级自然科学研究重点资助项目(KJ2012A257)

摘　　要：文章运用变点统计分析理论对上证指数连涨、连跌收益率进行Bayes实证分析。选用2005年6月6日至2010年5月12日共1 204个交易日的上证指数收益率数据,利用K-S检验法将其分成3个阶段,发现这3个阶段的连涨和连跌收益率服从伽马分布。通过计算后验概率的大小,判断分段之后的每一段收益率数据是否存在变点,同时给出变点发生位置的估计,并探讨变点发生的影响因素。T he statistic analysis theory of change point is used to analyze the successive rises and falls of returns of Shanghai stock index with the way of Bayes substantial evidence . The returns data of Shanghai stock index in 1 204 trading days from June 6 ,2005 to May 12 ,2010 are divided into three stages by K-S test ,and it is show n that the successive rises and falls of returns in the three stages yield to gamma distribution .Whether the change point exists in each stage is judged by the calculation of the Bayes posterior probability .Besides ,the estimation of the location of change points is given , and the factors influencing the occurrence of change points are discussed .

关 键 词：上证指数 连涨 连跌收益率 Bayes法 伽马分布 变点 

分 类 号：O212.8[理学—概率论与数理统计]