Re: [Gretl-users] Test for AR(2) errors: z = NA

On Tuesday, February 4, 2014 9:24 PM, Sven Schreiber <svetosch(a)gmx.net&gt; wrote: Am 04.02.2014 15:08, schrieb Anutechia Asongu: I don't really know what you mean by "information criterion for the test", but in Arellano-Bond the AR(1) test doesn't tell you anything that you didn't know already, first-order correlation is/should be in there by design. So, no. -sven _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users Thanks Sven, Well, I'm thinking of trying something else, limiting instrument proliferation by changing how the instrument is set and not by shrinking the sample (decreasing T and increasing N) as I did in my previous strategy. Can anyone gist me on the Gretl equivalent of Stata's xtabond2? http://www.cgdev.org/publication/note-theme-too-many-instruments-working-... Thank you in advance