Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: basic question about varlist


From   Roger Newson <roger.newson@kcl.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   st: Re: basic question about varlist
Date   Tue, 05 Nov 2002 20:30:32 +0000

At 15:00 05/11/02 -0500, Radu Ban wrote:
thanks a lot for the prompt replies. however, i think i didn't explain my
problem accurately.

i'll start from the end result perhaps it makes more sense this way. please
bear with me.
in the end, i would like to have a matrix (to be exported in excel) with
columns:

varname    coeff1    stderr1    coeff2    stderr2....

the variables under varname come from the varlist `outcomes' i described in
the previous email (i.e. local outcomes " an_abbreviated_varlist ").
to obtain the above mentioned matrix, i'm thinking of implementing the
following loop:

foreach var of varlist `outcomes' {
    reg `var' x1 x2...
    now somehow generate a scalar indexed by the position of `var' in
varlist (i.e. scalar1         for first variable, scalar2 for second), equal
to the coefficient of x1
    do the same for the stderror of x1, coeff of x2, and so on
    }

after the loop is done take these scalars and put them into the above
mentioned matrix. the part i don't know how to do is to generate those
scalars
If you want to export your proposed matrix to Excel, then the best way might be to create this matrix as a Stata data set, with 1 observation per model and variables -varname-, -coeff1-, -stderr1- etc., and then to export it to an Excel-compatible generic text spreadsheet using -outsheet-.

There is more than one way to create such a Stata data set. However, I personally would first use my packages -parmest- and -dsconcat- to create a data set with 1 obs for each parameter of each of the models, and data on the estimates, standard errors, confidence intervals, P-values, and other results. I would then use official Stata's -reshape- command to reshape this data set into a wider data set, with one observation per model rather than one observation per parameter per model.

To find out more about the use of -parmest- and -dsconcat-, Radu might like to look at my presentation "Creating plots and tables of estimation results using parmest and friends", downloadable from the Stata User Meetings website at http://ideas.repec.org/s/boc/usug02.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/




© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index