Empirical Analysis Based On Consumer Price Index Of Residents

The final price of all products and services in the community is based on the consumer price[1]survey.On the one hand,it is closely related to the people's life,on the other hand,it plays an important role in the national economy price system.It is an important index of economic analysis and decision,and it is also the important basis of the general price level and national economic accounting.This article mainly chooses the consumer price index which reflects the fluctuation of the price level to carry on the analysis.The first part is mainly to forecast the consumer price index of the residents,firstly,using the least squares estimation method and the Bayesian parameter estimation method based on the MCMC algorithm to establish the AR(2)model for the Chinese consumer Price index data in 1951-2012.Then,we use the two models to forecast the consumer price index of the 2013-2015 and compare the predicted results of the model based on Bayesian parameter estimation with those of the least squares estimator.The second part is the use of the consumer price index data of 31 provinces in2016 for factor analysis to find the main factors affecting the level of price changes.The third part is the SOM neural network clustering Analysis of the data and divides 31 provinces according to some similar characteristics.Research conclusions:(1)The predicted results of the model based on Bayesian parameter estimation are more accurate;(2)There are 4 main factors that affect the level of consumer price changes;(3)The analysis of artificial neural network shows that the level of consumer price change in 31 provinces is divided into five categories.