st: post estimation command after mlogit with svy: and mim:

[chmeyer <chmeyer@uni-greifswald.de>]

Dear Stata-list members,

I am searching for an option to compute a p value for a dummy-coded 
factor variable with a post-estimation command after mlogit with mim 
and svy prefix commands. I use Stata 11 and multiple imputation was 
done with ice. Svy was used to account for cluster sampling (two stage 
sampling).The outcome variable has three levels and the predictor 
variable "cat" has also three levels and was entered with two dummy 
variables. Without mim and svy the following did what I deserve:

mlogit outcome cat_2 cat_3 age, b(0) rr
testparm cat_2 cat_3 , equation(#2)
testparm cat_2 cat_3 , equation(#3)

With svy I receive plausible results from, but I do not know if  this 
is valid because the users manual did not mention that testparm is 
posible with svy.:

svy: mlogit outcome cat_2 cat_3 age, b(0) rr
testparm cat_2 cat_3 , equation(#2)
testparm cat_2 cat_3 , equation(#3)

With mim or mim and svy the equation option after testparm was not 
allowed. Are there any alternatives to do the job in this case?

Thank you for your attention. Any idea would be welcome!

Christian Meyer
University of Greifswald

*
*   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/