Re: [Gretl-users] Goldfeld-Quandt heteroscedasticity test: a funny script

This what I did before your fist message. It uses unnecessary sorting since I still not good with all options of 'smpl' I tested it by calling R-session and using gqtest{lmtest} Oleh P.S. Actually I was not interesting in GQ results but in good using of 'smpl' 21 грудня 2015, 20:59:11, від "Riccardo (Jack) Lucchetti" < r.lucchetti(a)univpm.it >: On Mon, 21 Dec 2015, oleg_komashko(a)ukr.net wrote: Maybe. Except that, now that I look at it again, it's all wrong! :D (the 2nd "ols" command should come straight after the "smpl" command, otherwise "den" and "df1" come from the full-sample ols) ------------------------------------------------------- Riccardo (Jack) Lucchetti Dipartimento di Scienze Economiche e Sociali (DiSES) Università Politecnica delle Marche (formerly known as Università di Ancona) r.lucchetti(a)univpm.it http://www2.econ.univpm.it/servizi/hpp/lucchetti -------------------------------------------------------