Re: st: Looping with regression controls

[Nick Cox <njcoxstata@gmail.com>]

Your two commands fit the same model. The order of predictors affects
only the presentation of results.

Nick

On Wed, Feb 29, 2012 at 10:51 PM, Ras Dondo <ras.dondo@yahoo.com> wrote:

> I am new to Stata and having difficulty fitting a
> model. I have one dependent variable RECOVER coded 0 or 1. I also have
> 10 independent categorical variables AGE, EDUCATION, RELIGION, MARITAL,
> SOC_CLASS, COUNTRY_BIRTH, EMPLOYMENT, YEARS_WORKED, NO_PARTNERS, and
> FAMILY_SIZE. I wish to model the effect of each sociodemographic
> variable on the dependent variable RECOVER whiles controlling for the
> remaining nine independent variables. First time round I wish to model
> the effect of AGE on RECOVER whiles controlling for the remaining nine
> variables as below.
>
>
> xi: logistic RECOVER i.AGE i.EDUCATION
> i.RELIGION i.MARITAL i.SOC_CLASS i.COUNTRY_BIRTH i.EMPLOYMENT
> i.YEARS_WORKED i.NO_PARTNERS i.FAMILY_SIZE
>
> Second time round I
> will wish to model the effect of EDUCATION on RECOVER whiles controlling
> for the remaining nine variables as below.
>
> xi: logistic RECOVER i.EDUCATION i.RELIGION i.MARITAL i.SOC_CLASS
> i.COUNTRY_BIRTH i.EMPLOYMENT i.YEARS_WORKED i.NO_PARTNERS i.FAMILY_SIZE i.AGE
>
>
> My
> question is instead of repeating the logistic command is it possible to
> write one command maybe using looping that will let me avoid the
> repetition?
>
*
*   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/