|
|  |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: xtabond2 and intercooled Stata 9.2
Kit--
I greatly appreciate your comments and suggestions. Although I haven't completely figured out all aspects of XTABOND2 yet, at least my computer doesn't crash! Thank you again for your time and assistance.
Anand
----- Original Message -----
From: Kit Baum <[email protected]>
Date: Tuesday, February 27, 2007 5:49 pm
Subject: st: xtabond2 and intercooled Stata 9.2
To: [email protected]
> Anand said:
>
> I am quite new to Stata and I am having trouble using the XTABOND2
> program.. either my computer crashes or gives me an error message.
>
> I used the following command...
>
> xtabond2 chasim L.chasim L.chdiff L.asiminflr eta, gmm(L.chasim)
> iv(L.chdiff L.asiminflr eta) nomata
>
> ... and I got the following error message.
>
> =========================
> Building GMM instruments.no room to add more variables due to width
>
>
> With 50 time periods xtabond2 will attempt to create a gazillion
> instruments, and no matter how many Gb of RAM you have, it will
> fail.
> Either limit the lags used in the gmm instruments with the
> suboption
> lag( ) or use the collapse option to reduce their number.
>
> More fundamentally, you must consider whether this data set fits
> the
> 'small T, large N' criteria that rationalize the DPD (Arellano-
> Bond)
> estimator. 50 is not that small, and this may not be the most
> sensible
> estimation technique in the case where T is a significant fraction
> of
> N.
>
>
> Kit Baum, Boston College Economics
> http://ideas.repec.org/e/pba1.html
> An Introduction to Modern Econometrics Using Stata:
> http://www.stata-press.com/books/imeus.html
>
> *
> * 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/
>
