Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

st: Marginal effects after mvprobit


From   <[email protected]>
To   <[email protected]>
Subject   st: Marginal effects after mvprobit
Date   Fri, 7 Mar 2014 15:48:34 +0000

Elin Vimefall <[email protected]> wants marginal effects after -mvprobit-. This question has been asked several times on the list before. (Please search the archives before posting! Also please heed the list FAQ by citing the provenance of program modules.)

To me, your request suggests that you need to think a bit more first about what you're asking for.  Precisely which marginal effects do you want? Marginal effects for this model could refer to joint probabilities, marginal probabilities, etc. etc., and the number of possible answers increases with the number of equations in your model.   For this reason, the accompanying -mvppredict- program only provides predictions of the probability of all-ones or all-zeros cases. (Note that LIMDEP makes a similar sort of judgement, but different if I recall.)

Given -mvpredict-, you can take up Steve Samuels' recommendations (and as discussed earlier on the list in response to similar queries). Generate predicted probabilities for individuals with particular sets of characteristics under different scenarios, and compare their averages.

In short, (i) you need to decide what you really want to do (there is no unique definition of a "marginal effect" for this sort of model), and (ii) you won't get answers in a canned fashion (for good reasons).

Depending on the size of your model, you may find the code in my article with Cappellari in Stata Journal 6(2) a quicker way to accomplish these same ends. Also check out -cmp-.

Stephen (author of -mvprobit- and -mvppredict-)
------------------
Stephen P. Jenkins [email protected]


Date: Thu, 6 Mar 2014 09:52:57 +0000
From: Elin Vimefall <[email protected]>
Subject: st: Marginal effects after mvprobit
Hi 
Is it possible to get the marginal effects after mvprobit?
My model looks like this:
mvprobit (y1 = x1 ... xn) (y2= x1...xn) (y3 = x1...xn) (y4= x1...xn), robust
I use the latest version of the program and I have stata13. 

I would really appreciate if anyone can help me with this!
//Elin Vimefall

Please access the attached hyperlink for an important electronic communications disclaimer: http://lse.ac.uk/emailDisclaimer

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index