Hi R-Community, I have an historical time series of power prices and would like to determine the daily and yearly seasonal component for forecast purposes. Unfortunately the functions "decompose" and "stl" do not provide an always identical repeating seasonal pattern which could be continued in the future. Why and is there a possibility to adjust a periodical function like s(t) = a1 sin(f1 t) + b1 cos(f1 t) + a2 sin(f2 t) + b2 cos(f2 t) + ... to a time series, if necessary with given frequencies fi. Much thanks in advance, Hagen Schmoeller -- Dipl.-Ing. Hagen K. Schm?ller Institut f?r Elektrische Anlagen und Energiewirtschaft, RWTH Aachen Schinkelstra?e 6, D-52056 Aachen, Germany Tel.: +49 (0)241 80-96734 Fax : +49 (0)241 80-92197 Hagen.Schmoeller at iaew.rwth-aachen.de