Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: RE: adding count variables


From   n j cox <n.j.cox@durham.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: adding count variables
Date   Thu, 05 Jul 2007 18:07:58 +0100

I think this means:

In SPSS you can

(a) stack two data sets, identified (say) by values 1 and 2
on some identifier, into one;

(b) have a count variable only for observations for which that
identifier is 1;

(c) carry out some procedure (e.g. Poisson regression)
for those same observations;

(d) assign the coefficients to the other part of the dataset.

Is this possible in Stata? Or do you need to do it treating
the two parts of the data somehow as separate data sets?

The answer is: Yes, you can do it in Stata. You can fit a model
using only a subset of observations and then -predict- out-of-
sample. You just need all the predictors used in the model
to be non-missing for observations for which you want predictions.

(As the translation implies, I had a hard time decoding your post.
I presume English is your first language, unless it is Gaelic.
A little more attention to spelling, capitalisation, punctuation and layout might increase your readership.)

Nick
n.j.cox@durham.ac.uk

Fitzmaurice, Ann E.

just a quick question
in spss can stack two data sets with an identid-ier 1 and 2, have a
count variable on some events but only in the the section of the file
(coded 1) and after the procedure (in this case poisson regression) ,
the coefficients are assigned to the other section of the data set, is
this possible in stata, or do you run againgst data set a and then use
the predict against dataset 2 (as two seperate data sets)

*
* 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