论文信息 - Local trend estimation and seasonal adjustment of economic and social time series (with discussion)

Local trend estimation and seasonal adjustment of economic and social time series (with discussion)

The first main purpose of this paper is to compare the performance of a number of methods of estimating local trends levels in a sample of 23 monthly economic and social time series. This is done for cases where data are available up to 0,1,2,...,6 months respectively beyond the time point for which the trend estimate is required. The paper's second main purpose is to investigate methods of improving the performance of the X-11 seasonal adjustment procedure. The treatment given to these problems is pragmatic and empirical rather than theoretical. The paper concludes by recommending the following simple unified procedure for both trend estimation and seasonal adjustment. Each time a new observation becomes available, forecast the next 12 monthly values of the series concerned by stepwise autoregression or some other appropriate method. Apply standard X-11 seasonal adjustment to the observed series augmented by the 12 forecast values. The trend and seasonally adjusted values given by the X-11 printout for the present and previous time points are the values required. It is shown that this simple procedure has a very satisfactory overall performance.

J. Durbin | P. Kenny

[1] D. Cox,et al. An Analysis of Transformations , 1964 .

[2] Julius Shiskin,et al. The X-11 variant of the census method II seasonal adjustment program , 1965 .

[3] Gwilym M. Jenkins,et al. Time series analysis, forecasting and control , 1972 .

[4] C. Granger,et al. Experience with Forecasting Univariate Time Series and the Combination of Forecasts , 1974 .

[5] M. J. Murphy,et al. Seasonal Adjustment Based on a Mixed Additive‐Multiplicative Model , 1975 .

[6] G. C. Tiao,et al. Decomposition of Seasonal Time Series: A Model for the Census X-11 Program , 1976 .

[7] R. Bhansali,et al. Some properties of the order of an autoregressive model selected by a generalization of Akaike∘s EPF criterion , 1977 .

[8] Chris Chatfield,et al. The Holt-Winters Forecasting Procedure , 1978 .

[9] Clive W. J. Granger,et al. Experience with using the Box-Cox transformation when forecasting economic time series , 1979 .

[10] David A. Pierce,et al. Data revisions with moving average seasonal adjustment procedures , 1980 .

[11] P. Newbold. A NOTE ON RELATIONS BETWEEN SEASONALLY ADJUSTED VARIABLES , 1980 .

[12] J. Burman. Seasonal Adjustment by Signal Extraction , 1980 .

[13] George E. P. Box,et al. Estimating current trend and growth rates in seasonal time series , 1981 .

[14] Kenneth F. Wallis,et al. Seasonal Adjustment and Revision of Current Data: Linear Filters for the X‐11 Method , 1982 .