Re: st: arellano-bond

["David M. Drukker" <ddrukker@stata.com>]

murali n.v <muralinv2001@yahoo.co.in> wrote that

> When I run the one-step method(see below for the
> code),I am getting the result without dropping any
> variable or dummies.But when I am doing the
> two-step(see below for code),it is droping most of the
> time dummies as well as the lagged dependent variable
> appearing in the RHS.
>
> ONE_STEP
> xtabond  lny  dum74 dum75 dum76 dum77 dum78 dum79
> dum80 dum81 dum82 dum83 dum84 dum85 dum86 dum87 dum88
> dum89 dum90 dum91 dum92 dum93 dum94 dum95  dum96 dum97
> dum98 dum99 , noconstant diffvars( lny70 sch70)
> lags(3) pre(pop gcf trade,  lagstruct(2,2)) robust
> artests(2)
>
> TWO-STEP
> xtabond  lny  dum74 dum75 dum76 dum77 dum78 dum79
> dum80 dum81 dum82 dum83 dum84 dum85 dum86 dum87 dum88
> dum89 dum90 dum91 dum92 dum93 dum94 dum95  dum96 dum97
> dum98 dum99 , diffvars( lny70 sch70) lags(3) pre(pop
> gcf trade,  lagstruct(2,2))  artests(2) twostep
I suspect that specifying the -noconstant- option in the one-step code but
not the two-step code is causing the differences.

	David
	ddrukker@stata.com

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/