<>
As an -ereturn- function, -e(sample)- is "ephemeral", though. If you want it
to stick around, turn it into a binary variable such as
-gen byte estsample1=e(sample)-
HTH
Martin
-----Ursprüngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von John Antonakis
Gesendet: Sonntag, 4. Oktober 2009 21:28
An: [email protected]
Betreff: Re: st: AW: Return results only from observations used in estimated
model
Dear Martin and Richard:
Great!
-if e(sample)-
......does the trick!
By table, I wanted to table the frequencies of the categorical
predictors, i.e.,
table y x z if e(sample), row col sc
Now I get the same number of observations from the table as I did in the
logistic.
Thanks much.
John.
____________________________________________________
Prof. John Antonakis
Associate Dean Faculty of Business and Economics
University of Lausanne
Internef #618
CH-1015 Lausanne-Dorigny
Switzerland
Tel ++41 (0)21 692-3438
Fax ++41 (0)21 692-3305
Faculty page:
http://www.hec.unil.ch/people/jantonakis&cl=en
Personal page:
http://www.hec.unil.ch/jantonakis
____________________________________________________
On 04.10.2009 21:21, Martin Weiss wrote:
> <>
>
> Just condition on -e(sample)-. I am not quite sure what "table the
results"
> is supposed to mean, though.
>
>
> *************
>
> vers 10.1
>
> clear*
> set obs 150
>
> gen x1=26+int(27*runiform())
> gen x2=runiform()<.48
>
> gen byte draw = /*
> */ runiform() < /*
> */ invlogit(1+.02*x1+.3*x2)
>
> logistic draw x*
>
> tempvar tem
> gen `tem'=runiform()
> sort `tem'
>
> drop in 121/150
>
> logistic draw x*
>
> table draw if e(sample), /*
> */ c(freq mean x1 mean x2)
> *************
>
>
>
> HTH
> Martin
>
> -----Ursprüngliche Nachricht-----
> Von: [email protected]
> [mailto:[email protected]] Im Auftrag von John
Antonakis
> Gesendet: Sonntag, 4. Oktober 2009 21:17
> An: [email protected]
> Betreff: st: Return results only from observations used in estimated
model
>
> Hi:
>
> Suppose I run a model using logistic on a data set with 150 observations:
>
> logistic y x z q r
>
> Now, suppose that because of listwise, the number of observations in the
> logistic model are 120.
>
> How would I table the results (using say only variable x and z) based
> ONLY on the data from the logistic model (i.e., the listwise sample) and
> not the whole sample ?
>
> Thanks in advance,
> John.
>
*
* 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/
*
* 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/
