Re: [Gretl-users] Test for AR(2) errors: z = NA
On Tue, 4 Feb 2014, Sven Schreiber wrote:
Am 04.02.2014 22:03, schrieb Anutechia Asongu:
>
>
> 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?
How to choose the (amount of) instruments is actually documented for the
dpanel command... RTFM (no offense intended)
See in particular the section titled "Instruments" in the Chapter on
dynamic panel models in the Gretl User's Guide.
Allin Cottrell