基于X-12-ARIMA调整的中国CPI波动检验  

China’s Fluctuation Test Based on Adjustment of X-12-ARIMA

作　　者：顾光同 Gu Guangtong(School of Economics and Management, Zhejiang A ＆ F University;Zhejiang Farmers Development Research Center, Hangzhou 311300, China)

机构地区：[1]浙江农林大学经济管理学院 [2]浙江省农民发展研究中心

出　　处：《统计与决策》2018年第11期18-22,共5页Statistics & Decision

基　　金：全国统计科学研究基金资助项目(2013LY123)

摘　　要：消费者物价指数（CPI）作为判断通货膨胀水平的指标,受到政策制定部门和学术界普遍关注。文章基于2001年1月至2015年10月CPI同比月度数据,通过描述性分析、广义波动检验、格点自助法和自回归模型的最小二乘估计法等对CPI和八大类子成分的统计特征、波动情况和惯性水平进行研究。结果表明,虽然中国CPI暂不存在显著结构性变化,但与八大类子成分的统计特征、波动情况和惯性水平存在显著的差异。因此,在制定CPI政策目标时需将不同子成分的不同特征纳入考虑范围,并及时更新统计分析方法。Consumer Price Index, as an indicator of judging the level of inflation, is widely concerned by policy-making departments and academia. Based on monthly CPI data from January 2001 to October 2015, this paper analyses CPI and the eight subcomponents＇ statistical characteristics, volatility and level of inertia by use of descriptive analysis, generalized fluctuation tests, grid-bootstrap and OLS estimate of autoregressive model. The results suggest that although there is no structural change in China＇s CPI data, it is significantly different from eight subcomponents in aspects of statistical characteristics, volatility and inertia level. Therefore, when setting CPI policy goals, we should take into account the different characteristics of the different subcomponents and timely update statistical analysis method.

关 键 词：CPI 结构性变化 X-12-ARIMA 广义波动检验 格点自助法 

分 类 号：C813[社会学—统计学]