摘要
本文首先讨论了计算中国实际经济时间序列的不同做法 ,并分析了其对季节调整的影响 ,指出通过同比增长率计算实际变量并进行季节调整是一个可以接受的做法 ,可以得到非常接近真实的季调后序列 ,并且在中国现有数据资源的限制下拥有一些特别的优势。然后本文具体讨论了对几个不同经济变量进行季节调整的方法 ,并给出了一些在经济数据分析与预测中的简单应用。方法的关键是采用regARIMA模型 ,从而可以对工作日变化、放长假、春节因素等作出一个估计和调整。作为一个副产品 ,本文引荐了一个相对较新的季节调整程序 (方法 ) ,TRAMO SEATS ,简单介绍了它的原理和优势 ,希望今后能得到更广泛的应用。
This paper discusses the different methods to calculate the real economic time series of China and analyze their effects on seasonal adjustment.With the above method,we can obtain the seasonally adjusted series which are very similar to the real ones.Then the paper discusses the specific seasonal adjustment methods to several economic variables and gives some simple applications in the analysis and forecasting of the economic data.The key of the method is to adopt the regARIMA model,in order to estimate and adjust the effects of variation in working days,long holidays,and Spring Festival,etc.The paper also introduces a newer seasonal adjustment procedure,TRAMO/SEATS,as a side product and presents its principles and advantages.It is hoped that the procedure can be widely used from now on.
出处
《经济研究》
CSSCI
北大核心
2002年第3期36-43,共8页
Economic Research Journal