德国BV4.1模型修正与中国CPI季节调整

发布时间：2018-03-26 07:32

本文选题：季节调整　切入点：BV.　出处：《统计研究》2017年07期

【摘要】：居民消费价格指数(CPI)是衡量通货膨胀程度和经济活动水平的重要指标,通常要剔除季节性因素影响。本文对国际最新的BV4.1季节调整模型进行了系统的研究和软件开发,编写R程序增强了其实用性。首先考虑到了中国的节日因素,交易日因素和异常值,对2001年1月至2015年3月的CPI数据进行了预处理。在分离出季节成分以及日历成分之后,采用平滑区间和修正历史法进行模型诊断的研究。研究认为:CPI的趋势在短期内具有二阶多项式发展特征,节日因素、交易日影响和异常值不显著;实证结果表明,BV4.1的季节调整结果与其他模型如X-12-ARIMA和TRAMO/SEATS相比具有很强的稳定性。
[Abstract]:The consumer price index (CPI) is an important index to measure the degree of inflation and the level of economic activity, which usually excludes the influence of seasonal factors. In this paper, the latest international seasonal adjustment model of BV4.1 has been systematically studied and software developed. Writing the R program enhances its practicability. First, we preprocess the CPI data from January 2001 to March 2015, taking into account the festival factors, trading day factors and abnormal values in China. Using smoothing interval and modified history method to study the model diagnosis. It is concluded that the trend of the component CPI has the characteristics of the second order polynomial development in the short term, the festival factor, the influence of trading day and the abnormal value are not significant. The empirical results show that the seasonally adjusted results of BV 4.1 are more stable than those of other models such as X-12-ARIMA and TRAMO/SEATS.
【作者单位】：暨南大学经济学院统计学系;中国社会科学院数量经济技术经济研究所;暨南大学经济学院;
【基金】：国家哲学社会科学基金项目“时间序列分解与中国经济下行压力下的风险识别及预警研究”(16BJY014) 教育部人文社会科学基金青年项目“中国居民消费价格指数数据质量优化与通货膨胀治理”(13YJC910004) 中央高校基本科研业务费“中国季节调整模型建构与环比增长率测算”(332201412614801)、“中国CPI数据质量新框架下的系统优化”(332201512615309)的资助
【分类号】：F224;F726

【参考文献】