El jue, 03-06-2010 a las 12:05 -0700, deepak bhatia escribió: The periodicity of the data is what defines the type of seasonal length. If you have monthly data, periodicity is 12 and the seasonal difference is y(t)-y(t-12) if your data are quarterly, periodicity is 4 and the seasonal difference is y(t)-y(t-4). Gretl admits periodicity 1 (annual), 4 (quarter), 12 (month) and 52 (week) having as reference period (for defining the seasonals cycles) the year, diary periodicity with 5 days, 6 days and 7 days, taking as reference the week, and hourly taking the day as reference. There is also the possibility of defining decennial data and user-defined periodicity. When you build your dataset with gretl you may specify the periodicity when gretl asks for the type of data (time series/cross section/panel) and you select time series. If you have imported your data from other source you can change the periodicity by selecting "Time series" in the menu "/Data/Dataset structure". -- Ignacio Diaz-Emparanza DEPARTAMENTO DE ECONOMÍA APLICADA III (ECONOMETRÍA Y ESTADÍSTICA) UPV/EHU Avda. Lehendakari Aguirre, 83 | 48015 BILBAO T.: +34 946013732 | F.: +34 946013754 www.ea3.ehu.es