统计研究

• 论文 • 上一篇    下一篇

季节调整新方法及其原理研究

何永涛 张晓峒   

  • 出版日期:2016-11-15 发布日期:2016-11-11

New Developments of Seasonal Adjustment and Theoretical Research

He Yongtao & Zhang Xiaotong   

  • Online:2016-11-15 Published:2016-11-11

摘要: 本文的主要工作是从频域的角度对季节调整中“季节滤子”的设计及估计问题进行研究。通过将直接信号提取(DSEF)方法引入到季节调整的应用之中,突破现有季节调整方法中仅能处理季度或月度数据的限制,且该方法下季节调整后的序列是理论季节调整后序列的“均方误差”最小估计。将DSEF方法应用于对中国季度进出口总额序列的季节调整分析中。分析结果显示,相比于X-11和SEATS方法,DSEF方法季节调整结果的离差较小且稳健性较好。

关键词: 季节调整, X-11, 信号提取, TRAMO-SEATS

Abstract: In this paper, the program about seasonal filters’ design and estimation based on frequency domain are studied. A new seasonal adjustment method is presented via employing a direct signal extraction theory, which breaks through current SA methods’ limitation that can only deal with seasonal or monthly dates, and the output of the new SA method is the minimum in MSE. Finally, we apply this new method to analyze the monthly series of China’s import and export. The results show that the method we presented has a smaller deviation and better robustness, comparing with X-11 and SEATS method.

Key words: Seasonal Adjustment, X-11, Signal Extraction, TRAMO-SEATS