RE: st: Problem returning the e(V) covariance matrix for looped regression

[Sharon Walsh <Sharon.Walsh@esri.ie>]

Hi Nick,

Many thanks for your help!! I tried this method and it seems to work perfectly except that it only runs for the first 20 or so countries (out of a total of 246) and then I get the following error:

no observations
r(2000);

end of do-file

r(2000);

Do you have any thoughts?

Many thanks again for your help so far,
Kind regards,
Sharon


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nick Cox
Sent: 10 October 2011 11:46
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: Problem returning the e(V) covariance matrix for looped regression

You need to loop over your countries saving the matrix each time. Here
is some technique:

sysuse auto
egen group = group(rep78)
su group

forval i = 1/`r(max)' {
      regress mpg weight if group == `i'
      mat V`i' = e(V)
      mat li V`i'
}

Also -search levelsof-.

Nick


On Mon, Oct 10, 2011 at 11:37 AM, sharoncatherine <sharon.walsh@esri.ie> wrote:

> I have StataSE 11 for Windows. I have sorted my data by country of origin
> and I ran an OLS regression *by* country. I now need to get the covariance
> matrix for each country, however, when I enter the command matrix list e(V)
> it only returns the most recently run regression which in this instance is
> the last country in the list. I would really appreciate if someone could
> guide me as to how to get the covariance matrix for each country run.
>
> Any help would be greatly appreciated!!
>
>
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/