Statalist


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

st: suest with multinomial logit


From   daphna bassok <dbassok@stanford.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: suest with multinomial logit
Date   Thu, 10 Jul 2008 23:10:39 -0700


Hi-

I am tying to use the "suest" command for the first time, and I wanted to make sure that I am doing so appropriately.

I am running a multinomial logit model predicting child care choice. I run the exact same model separately for several racial groups (white, black, Hispanic).

I'd like to do two things...

1, The first is compare coefficients across the different models. Is it appropriate to use suest this way?

Here is my code (after running mlogit & storing estimates):

# delimit ;
local estimates "black_est hisp_e_est hisp_ne_est " ;
foreach x of local estimates { ;
foreach var of varlist var1-var10 { ;
test [white_est_center]`var' - [`x'_center]`var' = 0 ;
} ;
} ;

I believe this is testing whether each of the coefficient is significantly different between the white model and another group... when comparing the "center" option to the omitted option. I then repeat this procedure for the other choices in the multinomial model.

2. Is there a way to use suest to simultaneously test whether the full set of coefficients in the white model are significantly different from the full set of coefficients for say the black model?

Thanks in advance for any advice you can provide.

Daphna

*
*   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