<>
Which version of Stata do you use? Seems to me that -ssc d desmat-`s
functionality has been pretty much rendered redundant by the new factor
variables (-h fvvarlist-) in Stata 11. Note you are supposed to let the list
know if using anything <11...
Anyway, the problem you complained about also occurs with the most prominent
Stata dataset:
*************
sysuse auto, clear
gen svyweight=abs(rnormal())
svyset _n [pweight=svyweight], vce(linearized) singleunit(missing)
svy: reg rep78 price mpg headroom
desmat:svy: reg rep78 price foreign*@headroom, verbose
*************
HTH
Martin
-----Ursprüngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von Björn Bünger
Gesendet: Mittwoch, 17. Februar 2010 13:28
An: [email protected]
Betreff: st: using desmat svy reg with dependent variable _x_1 ?
Dear statalisters,
I want to estimate a regression with the following syntax:
desmat: svy: reg sathlth obesedum*@docvis
Two problems arise: First, the regression is reported using "_X_1" as
dependent variable (which I suppose means that the first category of
(ordinally scaled) sathlth is used as dependent variable.)
The second (and in my opinion more severe) problem is that the other
catogories for sathlth are included in the output as independent variables.
If I define sathlth as continuous variable (@sathlth) this effect does not
occur, but even if I estimate the equation in an ordered probit framework it
does (desmat: svy: oprobit sathlth obesedum*@docvis ).
How should I proceed if I want to include sathlth only as dependent
variable?
I'm grateful for advices and hints
Björn Bünger
Institute of Public Economics
University of Münster
Germany
*
* 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/
