基于VAR模型的CPI与PPI因果关系检验

Research on the causality relation between CPI and PPI:A VAR perspective

选取国家统计局1996年10月—2019年2月公布的CPI和PPI月度同比数据,通过构建向量自回归(VAR)模型,运用格兰杰因果关系检验、脉冲响应函数以及方差分解法,研究CPI与PPI之间的动态关系.研究结果表明:在选用的滞后期为2的VAR模型中,CPI与PPI互为因果关系,它们之间保持着相互影响的正向时滞关系;CPI与PPI受到自身的冲击影响较大,而且二者对自身有正向的影响;CPI的变动90%由自身引起,10%由PPI引起,PPI的变动80%由自身引起,20%由CPI引起.因此,在研究CPI与PPI的变动问题时,应重点关注自身变动,同时注意相互影响.

The paper establishes a VAR mode,which uses the monthly data of CPI and PPI from October 1967 to February 2019,to study dynamic relationship between CPI and PPI.The data was downloaded on the National Bureau of Statistics.In order to analyze the VAR model,this paper uses a Granger causality test,impulse response function and variance decomposition method.The results show that CPI and PPI are causal in each other in the VAR model with lag period of 2,and they maintain a positive time-lag relationship with each other.CPI and PPI are greatly affected by their own,and both have a positive impact on themselves.90%change of CPI is caused by itself,and 10%by PPI.80%of the change in PPI is caused by itself,and 20%is caused by CPI,so it is important to focus on their own changes and mutual influence when studying CPI and PPI.