Hello all, Apologies to those of you who get this message twice. But Jack and I having been working (on and off) since June on a new dynamic panel-data estimation command, "dpanel", which we are now ready to unveil. It's documented in dpanel.pdf at http://sourceforge.net/projects/gretl/files/manual/ We think it's a pretty nice implementation of the GMM-DIF (Arellano-Bond) and GMM-SYS (Blundell-Bond) estimators, but we'd be very grateful to get any feedback you may have before we put out a release in which this command is truly visible (it has been there "secretly" for a while now). Since we have a full ChangeLog, ideally we'd like to make a new release next week. If you can spare some time to test dpanel before then, that would be great. We're looking for any show-stopping bugs and/or severe differences over what the default behaviour of this command should be. We have tried to build in what we take to be sensible defaults so that "standard" invocations of the command are short and sweet, while more complex or non-standard uses are also supported. Note that at this point there's no GUI access to dpanel, it's a script-only thing. If we have the script command right, a graphical interface may follow. [Footnote: Sven Schrieber suggested that this command might better be named "dynpanel", and we invited votes on this on the gretl-devel list. But, the vote being indecisive (only Ignacio Diaz-Emparanza put in an opinion), I'm afraid that the preference of Jack and myself for brevity has prevailed!] Allin -- Allin Cottrell Department of Economics Wake Forest University _______________________________________________ Gretl-devel mailing list Gretl-devel(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-devel