Peter,
In full agreement with Rafa. Note that if you would like to account for
differences between full time and part time workers among the other
variables in the data set, you might be able to apply the xi command and
generate interaction variables between the full time worker indicator
variable and the other variables of interest which may vary by full time
vs. part time worker status.
Along the lines of what Rafa is saying below, this will allow you to keep
the full data set and better assess differences by worker status.
Joe Dorris
"R.E. De Hoyos"
<[email protected]
om> To
Sent by: <[email protected]>
owner-statalist@h cc
sphsun2.harvard.e
du Subject
st: Re: Full-time workers only
regression
05/31/2006 02:53
PM
Please respond to
statalist@hsphsun
2.harvard.edu
Peter,
The two options that you mentioned are quite different.
If you run "y" on the full data including a dummy for full-time labourers,
then you would be assuming that the only difference between full-time
labourers and others is given by the dummy. The parameters for all other
variables included in the model are restricted to be the same across labour
market categories. For example, if one of your regressors is the years of
schooling, then the estimated parameter \beta_{schooling} would tell you
that the returns to schooling are the same regardless of the labourer being
a part or full-time worker.
You second option (restrict estimation to full-time workers) implies that
the model is only valid for full-time workers. You wouldn't be able to say
anything about part-time workers.
I hope this helps,
Rafa
________________________
Rafael E. De Hoyos
Faculty of Economics
University of Cambridge
CB3 9DE, UK
www.econ.cam.ac.uk/phd/red29/
----- Original Message -----
From: "peter harper" <[email protected]>
To: <[email protected]>
Sent: Wednesday, May 31, 2006 7:33 PM
Subject: st: Full-time workers only regression
> Dear Statalist
>
> This is a simple question. I am using a cross-section data with different
> firms and workers. I want to run a regression on full-time workers only.
> Would it be best to put the variable, say, flprt in the regression which
> indicates whether the worker works full or part-time and using the dummy
> variable for full-time, frlprt2; where frlprt2=full-time workers.
>
> (2) Or run
>
> y x1 x2 x3 if flprt==2
>
> where frlprt2=full-time workers.
>
> Thanks
>
> *
> * 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/
>
*
* 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/
*
* 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/
