Chain index constructed with seasonal adjustment.pdf

Vol.7 No.4 (2010) p79-85 Journal of Management Science Statistical Decision CHAIN INDEX CONSTRUCTED WITH SEASONAL ADJUSTMENT DAN-HUI YI AND HONG-WEI XIAO School of Statistics Renmin University of China Beijing, 100872, China xiaohongwei2006@126.com ABSTRACT. This paper, based on time series seasonal adjustment, proposes the construction method of statistics chain index, firstly decomposes the time series structure, according to the proportion of each component in different periods, to determine the dynamic weight, then calculates statistical chain index of each component, finally synthesizes the statistical chain index. When adding a new monthly data, we do seasonally adjustment again, and adjust the chain of previous statistics index accordingly. We use this chain index construction method on China monthly exports data, result shows that statistical chain index is stabile and easy to compare between different months. Keywords: Seasonal Adjustment; X-12-ARIMA; Statistical Chain Index 1. Introduction. China announced economic time series data is divided into two categories: the amount of data, year-on-year data. In order to eliminate seasonal factors, Chinas monthly statistics mainly use year-on-year index, which compares the two periods’ raw data directly, now the main announced data is the last year over the same month of the year index. The year-on-year index facilitates the calculation, but due to the interval of two comparisons is 12-month, as well as by the impact of hikes, in economic time series change, as an indicator of turning points in response to economic change, year-on-year index responses more slowly than the chain index, and is not conducive to receive rapid feedback from changes, affecting the timeliness and accuracy of decision-making. Therefore, in order to improve the timeliness of statistics, it is necessary for the calculation of statistical chain index. When calculating the chain index, due to seaso