Derar all, This is a problem that I am unable to solve. I am interested in analyzing some high-frequency data which show two periodicities (periods 5 and 260), It is clear how to estimate models with one periodicity with arima() but I have not been able to include both periodicities. Is there some good way to handle tis kind of problem? Thanks in advane, Miguel A. -- ************************* Miguel A. Arranz Tol-Project maarranz at tol-project.org