Change Point Identification Problem Of Normal Distribution Sequence

ASAMC algorithm is applied in this article to solve the identification of a normal distribution sequence change point problem,we have achieved the identification of both change point number and locations,and through the stochastic simulation,the efficiency of the identification accuracy is verified.Finally,this method was applied to the annual average temperature series and monthly average temperature series in Urumqi City from 1973 to 2017 for 45 years,and the temperature change point was found successfully.This paper is divided into five parts.the first part mainly introduces the significance of the research background of the change point problem,the research status of domestic and foreign scholars and the brief introduction of Bayesian statistical inference.In the second part,we first introduce the model of the normal distribution and give the prior distribution of each parameter,then we get the posterior distribution function of the change points.In the third part,the M-H algorithm and ASAMC algorithm are introduced.Then ASAMC algorithm is applied to the minimum posterior probability of the negative logarithm of the change point and the concrete steps are given.In the fourth part,the above theoretical knowledge is verified by random simulation,and the simulation results show that the identification of the number and position of change points is very accurate.In the fifth part,the method is applied to the average temperature series of Urumqi from 1973 to 2017,and the temperature change points are found.