Re: st: how to make xi dummies inherit labels

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

At 10:29 04/10/02 +0100, Roger Harbord wrote:
> --On 04 October 2002 04:07 -0500 Fred Wolfe 
> <fwolfe@arthritis-research.org> wrote:
>
>
> > At 09:57 AM 10/4/2002 +0100, you wrote:
> > > If version 1.1.3 is available from sealedenvelope.com there's something
> > > wrong with the way it's interfacing with -net- :
> > You can also get the most up to date version by ssc install reformat,
> > replace. I just did.
> >
> > ! version 1.1.3  26Aug02 arb
> > *! Reformats the output from regression commands
> > *! (c) Sealed Envelope Ltd, 2002
>
> Then why doesn't it work for me???  :
For what it's worth, an alternative way forward for Roger might possibly be 
to use my packages -parmest-,  -factext- and -descsave-, all downloadable 
from SSC. The -parmest- package creates post-estimation data sets with 1 
observation per model parameter (or per parameter per by-group) and data on 
estimates, confidence limits and P-values. And, if some of these parameters 
belong to -xi- dummies, then it is possible to reconstruct the variable 
from which the dummies were generated (complete with value labels) using 
the -descsave- and -factext- packages. For instance, if, in the -auto- 
data, we define a variable -country- indicating a car's country of origin, 
then we can type

tempfile tf1
descsave country,do(`tf1',replace)
parmby "xi:regress mpg i.country",label norestore
factext,do(`tf1')

and create a data set in the memory with 1 observation per model parameter, 
containing a variable -country- (with value labels from the original 
variable) denoting the country corresponding to the parameter.

I gave a presentation about how to do this at the 2002 UK Stata User 
Meeting, and it can be downloaded from 
http://ideas.uqam.ca/ideas/data/bocusug02.html.

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

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/