CPI链式漂移主因识别及改进方法研究

A Study on the Main Cause Identification and Improvement Method of CPI Chain Drift

利用大数据创新编制居民消费价格指数(CPI),已成为各国政府和学术界共同关注的焦点。但在高频大数据背景下,链式价格指数的漂移问题是一个尚未有效解决的难题。对此,本文基于链式漂移主要影响因素提出了程序化构造抗漂移灵活商品篮子的对策方法。利用京东商城大数据进行试算的实证研究表明:首先,少数商品交易价格的大幅弹跳,是引发链式价格指数严重漂移的主要原因;其次,程序化识别并剔除引发链式漂移的商品项目后,据此构造的灵活商品篮子具有较好的代表性,相应Tornqvist链式价格指数显示出明显的抗漂移效果和较准确的预测能力;最后,本文提出的抗漂移灵活商品篮子构造方法具有较强的可操作性,为化解高频大数据链式价格指数漂移提供了新思路,同时也将为相关指数的实践编制提供参考。

The innovative compiling of consumer price index by using big data has become a hot issue for governments and the academic community all over the world.However,the drift of chain price index is a difficult problem that has not been effectively solved with the sources of high-frequency big data.In this paper,a method of constructing anti-drift flexible commodity basket based on the main influencing factor of chain drift is proposed to solve this problem.Based on the big data of JD platform for the trial calculation,empirical research shows:First of all,the abnormal price bounce of a few commodities is the main cause of the serious CPI chain drift.Secondly,through the procedural identification and removal of commodity items that cause the drift,the constructed flexible commodity basket has a good representation and the corresponding Tornqvist chain price index shows obvious anti-drift effect and more accurate prediction ability.Finally,the method presented in this paper has good practicability,which provides a new idea to resolve the drift of chain price index from high-frequency big data,and also provides some reference for the practical compilation of relevant indexes.