I am having a strange problem with margeff and the dummies list that I
don't quite understand, and that I've replicated with other data. I
originally wanted to run:
svy: probit female height weight dd1 dd2 dd3 dd4 dd5 dd6 dd7 dd8 dd9
dd10 dd11 dd12 dd13;
margeff, count percent dummies(dd1 dd2 dd3 dd4 dd5 dd6 dd7 dd8 dd9
dd10 dd11 dd12 dd13);
This produced the "estimates post: matrix has missing values" error.
After looking at the trace and fiddling with the -dummies- list, I
came to the conclusion that there was some sort of problem with dd1
and dd11. When I renamed dd1 to something else it works just fine. Is
there a way to get around this? I saw that there was an earlier
problem with abbreviations that resulted in a similar error.
Here's the code:
#delimit;
use "http://www.stata-press.com/data/r9/nhanes2f.dta";, clear;
qui tab houssiz, gen(dd);
// dd14 is omitted in all specifications
which margeff;
/* Replace dd1 with dd01 */
gen dd01=dd1;
svy: probit female height weight dd01 dd2 dd3 dd4 dd5 dd6 dd7 dd8 dd9
dd10 dd11 dd12 dd13;
margeff, count percent dummies(dd01 dd2 dd3 dd4 dd5 dd6 dd7 dd8 dd9
dd10 dd11 dd12 dd13);
/* Intended Specification */
svy: probit female height weight dd1 dd2 dd3 dd4 dd5 dd6 dd7 dd8 dd9
dd10 dd11 dd12 dd13;
margeff, count percent dummies(dd1 dd2 dd3 dd4 dd5 dd6 dd7 dd8 dd9
dd10 dd11 dd12 dd13);
*
* 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/
