Research On The Compilation Of New Consumer Price Index

Consumer price index(CPI)is a price index compiled by the statistical department of our government,and it is an important indicator to measure economic development.It is usually used to measure the price level changes of commodities purchased by households in consumer products and services,which is conducive to understanding and analyzing the formulation of China’s economic policies and the basic trends of market prices.However,there are still some problems in the process of CPI collection and compilation,which reduces the timeliness and accuracy of the index.Nowadays,with the rapid development of Internet and e-commerce,there are more and more data and information.Due to the advantages of convenience,affordability and efficiency,more and more people choose online shopping.At the same time,the price of electronic goods more truly reflects the daily consumption level of residents.In this case,the Tsinghua University project team released the online consumer price index(i CPI),which had a great impact with its unique advantages.However,because i CPI has just come out,its representativeness and authority are still in doubt.Therefore,how to effectively combine i CPI and CPI to improve the problems existing in China’s price index has become a hot topic of concern to most scholars recently.Therefore,understanding the relationship between traditional CPI and i CPI,and getting a brand-new statistical index through the information of network data to improve the lag of traditional CPI index,so as to better reflect the rapid changes of commodity prices,is the research focus of this article.Firstly,this paper analyzes the relationship between them by means of stationarity test,cointegration test and Granger causality.It is found that there is a long-term close relationship between i CPI and CPI,and the Granger causality results show that i CPI has a certain lead for CPI.Then,through HP filtering,cross-spectrum analysis and other methods,the specific values of CPI and i CPI leading and lagging are obtained.The results show that except for clothing and medical care,the changes of other seven categories are ahead of CPI,which shows that the changes of online price index are faster than those of official CPI,and it is feasible to improve CPI index through i CPI data.Then the weighted average is used to correct the CPI in the same period to construct a new CPI index,namely nCPI.This makes the revised data more truly reflect the influence of the online price index on the traditional price index,which is more convincing.Finally,nCPI,CPI and i CPI are compared by empirical mode decomposition,HP filtering and cross-sectional analysis,and the advantages of nCPI are obtained.The results show that there are great differences betweenCPI and i CPI in both long-term trend and short-term change.To some extent,this difference means that i CPI can provide new information for improving CPI index.Finally,in the comparative analysis of the three indexes,it is found that nCPI has a higher correlation with CPI,a shorter lead and lag time,and a better correction effect,which can better make up for the shortcomings of CPI and i CPI.With the increasing popularity of online sales and the trend of online development of the real economy,the role of nCPI will certainly become more and more obvious.