Dear gretl users, The command "fcast" is not enought documented . Is it possible to get more informations and formulas for - dynamic option : is it possible to link for the source code (cc+) ? I have not the book :Davidson, R. and J. G. MacKinnon (2004) Econometric Theory and Methods, New York: Oxford University Press. Perhaps more informations about formulas to put in the "Aide, help" An I have an other reference (Pagan, Nicholls (1984)...) - and for an AR1 model ?: is it easy to compute the forecast (by repeated substitution , the prediction requires knowledge of XT+s but not of the values of X in the periods between T and T+s) But which option (s) for the forecast standard error ?.Further details ?( Judg et al (1985), Harvey (1990) ,..?..) More help about these points.Please... Sincerely and thanks Michel POUCHAIN