Kaiser Remiro, ReginaMaravall, AgustínUniversidad Carlos III de Madrid. Departamento de Estadística2011-01-182011-01-182000http://hdl.handle.net/10016/10010The paper details an application of programs. TRAMO and SEATS to seasonal adjustment and trend-cycle estimation. The series considered is the German Retail Trade Turnover series, for which, when adjusting with XI2-ARIMA, the Bundesbank had identified two problems. One had to do with heteroscedasticity in the seasonal component, associated with very different moving patterns for some of the months. The other one was related to the stability of the trend-cycle at the end of the series. It is seen how, starting with the fully automatic procedure and adding some simple modifications, the ARIMA-model-based approach of TRAMO-SEATS deals properly with both problems and provides good results, that are stable and robust.application/octet-streamapplication/octet-streamapplication/pdfengAtribución-NoComercial-SinDerivadas 3.0 EspañaSeasonal adjustmentPreadjustment stabilityAutomatic procedureworking paperEstadísticaopen access