Model-based seasonal adjustment implicitly defines a set of weights at
the ends of series as well as in the middle. Until now, with the exce
ption of very simple models, the weights have been obtained numericall
y. In this paper we give the analytical expressions for the weights fo
r both the structural and the ARIMA framework for a model which contai
ns trend, seasonal and irregular component. In the final part of the p
aper we address the question of robustness of model-based seasonal adj
ustment. We analyse practically, using real time series, and theoretic
ally, through the analysis of the shape of the weights, how the fittin
g of different specifications for the non-seasonal part affects the ex
traction of the seasonal component. (C) 1998 John Wiley & Sons, Ltd.