Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: SE AND CI FOR OF COMBINATIONS OF MARGINAL EFFECTS


From   Giannis Paraskeuopouls <gparaskeuopouls@hotmail.gr>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: SE AND CI FOR OF COMBINATIONS OF MARGINAL EFFECTS
Date   Wed, 11 Aug 2010 19:58:18 +0100

<>

Dear all, 

I want to run 3 separate models. Indeed,

reg y x1 x2 x3 x4 x5 x6  [model y]

mlogit x2 x3 .....       [model x2]

probit x4 x3 ...         [model x4]

Then
 I am interesting in calculating non-linear combinations of the 
parameters of the different models. To be more precise, I would like to 
calculate combinations of the marginal effects (from the coefficients of
 the mlogit models) and the OLS coefficients.

Something like:

[y]_b[x4]*[x4]_marginaleffect(x3).

I
 try to use -suest command after the estimation of the 3 models. This is
 a good idea because it is doing what I want (I can do non-linear 
combinations from the different models), however, due to the fact that I
 am interesting in the marginal effects of the last models, -suest fails
 to save marginal effects from these models.

Of course, I can 
calculate the combinations by hand (by multiplying the co-efficient from
 the different models) but I cannot take SE or CI.

I really appreciate your help.



 		 	   		  
*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index