运用RJMCMC方法对上证指数连涨连跌收益率的Bayes分析  

Bayes Analysis of the Successive Rises and Falls of SSECI Yield by RJMCMC Method

作　　者：程慧慧 许淑月 CHENG Huihui;XU Shuyue(School of Mathematics and Statistics,North China University of Water Resources and Electric Power,Zhengzhou,China 452370)

机构地区：[1]华北水利水电大学数学与统计学院,河南郑州452370

出　　处：《温州大学学报（自然科学版）》2023年第4期13-24,共12页Journal of Wenzhou University(Natural Science Edition)

摘　　要：运用RJMCMC方法对上证指数的连涨连跌收益率进行Bayes实证分析.选用2016年5月3日至2022年5月20日的上证指数收益率,在连涨和连跌股指服从伽马分布的基础上建立伽马分布双参数多变点模型,由Bayes分层理论得到参数后验分布的核密度函数;然后设计五种移动类型,分别给出其接受概率的表达式,建立RJMCMC算法;最后由算法确定连涨和连跌收益率存在的变点个数,给出变点位置的估计,并分析变点发生的影响因素.实证分析的结果表明,运用RJMCMC理论寻找股市变点的方法是有效的.This paper analyzes the successive rises and falls of SSECI yield by Reversible Jump Markov Chain Monte Carlo(RJMCMC)method.It selects the return rate of SSECI from May 3,2016 to May 20,2022,builds the change point model of Gamma distribution double parameters on the basis of successive rises and falls yield series following the gamma distribution,and obtains the kernel density function of posterior distribution of parameters by the hierarchical Bayesian method.Then five types of jumps are designed and the acceptance probability of each type is given to establish the RJMCMC algorithm.Finally,RJMCMC algorithm is used for determining the number of change points of successive rises and falls yield series,giving the estimation of the position of the change points,and analyzing the influencing factors of the change points.The analysis results show that the RJMCMC based method is efficient in finding the stock market changes.

关 键 词：上证指数 涨连跌收益率 伽马分布 变点 RJMCMC方法 BAYES分析 

分 类 号：O212[理学—概率论与数理统计]