st: Re: retaining residuals using "by group" with regression

[Roger Newson <roger.newson@kcl.ac.uk>]

At 06:49 27/01/04 -0800, David Bernstein wrote:
> Hello STATers,
>
> How do I retain my predicted values from a regression
> when using the "by group" command?
>
> for example:
>
> by company: reg x y1 y2 y3
> predict zresid, rstudent
>
> Is it possible to create a data file that contains the
> zresids that is associated with the regression for
> each each company?
I think the best way is probably in a loop, using the official Stata 
-levels- command and my own -dsconcat- package (downloadable from SSC), 
similarly to the following (assuming that -company- is an integer-valued 
variable):

levels company
local coms "`r(levels)'"
preserve
local flist ""
local i1=0
foreach com of numlist `coms' {
  local i1=`i1'+1
  tempfile tf`i1'
  restore, preserve
  keep if company==`com'
  regress x y1 y2 y3
  predict zresid, rstudent
  save `tf`i1'', replace
  local flist "`flist' `tf`i1''"
}
dsconcat `flist'

This should cause the memory to contain a data set with the existing 
variables plus a new variable -zresid- containing studentised residuals. In 
the more general case, where by-groups may be based on multiple variables, 
you can use the official Stata -gsort- command with the -generate()- 
option. And, if you also want to save the model parameters, then you might 
use the -parmest- package, also downloadable from SSC.

I hope this helps.

Roger


--
Roger Newson
Lecturer in Medical Statistics
Department of Public Health Sciences
King's College London
5th Floor, Capital House
42 Weston Street
London SE1 3QD
United Kingdom

Tel: 020 7848 6648 International +44 20 7848 6648
Fax: 020 7848 6620 International +44 20 7848 6620
  or 020 7848 6605 International +44 20 7848 6605
Email: roger.newson@kcl.ac.uk
Website: http://www.kcl-phs.org.uk/rogernewson

Opinions expressed are those of the author, not the institution.

*
*   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/