基于扫描数据的CPI基本分类指数编制方法研究

扫描数据是一种新的统计数据来源，在消费者价格指数编制中拥有很大的优势。国外统计部门从一开始就非常重视对这一新的数据来源的应用研究和实践。本文分析了荷兰、挪威、瑞士和瑞典四国目前应用扫描数据编制CPI的情况，选择奶酪和啤酒两个基本分类,分别运用Jevons和T?rnqvist公式编制CPI基本分类指数，编制结果与RYGEKS基准指数进行比较分析，分析显示：Jevons链指数与RYGEKS指数比较有下行偏差，且偏差较大；T?rnqvist链指数与RYGEKS指数比较偏差方向不明但差异较小。偏差的上行或下行不仅与公式的选择有关，还与产品的性质有关。根据国外的经验和本文的研究结果，提出我国应用扫描数据编制CPI的研究方向和建议。

Study on Compilation Method of CPI Elementary Classification Indexes Based on Scanner Data

As a new source of statistical data, scanner data have great advantages in the compilation of CPI. Foreign statistical departments have attached great importance to the study and application of this new source of data from the very first. In this paper, the experience of four countries using scanner data to compile CPI is studied, including Netherland, Norway, Switzerland and Sweden. Besides, this paper selects the cheese and beer groups as the research objects and compiles the elementary aggregates indexes using the Jevons and T?rnqvist equations，and the compilation results are compared with the RYGEKS which is regarded as benchmark index. The results are as follows: monthly chained Jevons price index has a downward large bias comparing with RYGEKS index, while monthly chained T?rnqvist index has a relatively small deviations comparing with RYGEKS in uncertain direction. Whether the bias is downward or upward relates to not only the choice of compiling formula, but also the characteristics of the products. In addition, according to the developed countries’ compilation experience and the findings in this paper, some recommendations are made about compiling CPI with scanner data in China.