|
|  |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Marginal Effects
At 01:59 PM 10/19/2006, Michael Blasnik wrote:
mfx needs to make predictions from the model that was fit by the
estimation command. The predictions from a command usually come
from a second ado file command_p. In this instance, Stata is
looking for dhurd_p to generate model predictions. You can try to
write such an ado file yourself (based on looking at similar
files). You may be able to use margeff (findit margeff), but I'm
not sure if that addresses this issue.
Yes, my guess is that dhurd_p.ado is one of the great unwritten
programs right now. dhurdle.ado includes the line
est local predict "dhurd_p";
You could try deleting that, and then the generic predict command
would get used instead. But, it may not be adequate for this
task. Customized predict commands will do things like predict
probabilities and other things that are semi-unique for the command.
Joe Hilbe has written several hurdle commands, available from
SSC. Michael W. might check to see whether any of them better fits his needs.
I wonder how many user-written estimation commands include customized
predict commands? It is very useful to have them, but it isn't
always a trivial matter to do so.
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX: (574)288-4373
HOME: (574)289-5227
EMAIL: [email protected]
WWW (personal): http://www.nd.edu/~rwilliam
WWW (department): http://www.nd.edu/~soc
*
* 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/
