The Choice of Time Interval in Seasonal Adjustment: Characterization and Tools

A typical problem of the seasonal adjustment procedures arises when the observed series is subject to structural breaks. In fact, using the full time interval, the seasonal adjusted series can differ from the "true" seasonal adjusted series, with unclear evidence showed by the usual diagnostic tests. Often the researcher has to decide where cut-off the observed series to obtain a homogeneous span; this is generally performed by a simple visual inspection studies of the graph of the series. In this paper we propose a statistical criterion based on a distance measure between filters, evaluating its performance with Monte Carlo experiments.