Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: predictnl for function across observations?


From   Gabriel Demombynes <[email protected]>
To   [email protected]
Subject   st: predictnl for function across observations?
Date   Tue, 4 Jan 2005 02:44:28 -0800 (PST)

I'm using predictnl to get out-of-sample predicted
probabilities and the associated variance using
estimates from an xtprobit regression. 

I am calculating the mean of these predicted
probabilities over by groups (each of which has 10
observations). I would like to calculate the standard
errors of these means. However, predictnl only gives
me the standard error (or variance) of the point
estimate, not the covariances between the predicted
values. I need the covariances to calculate the
variance of the mean.

Is there some way to calculate what I want easily in
Stata? Could nlcom do this? The alternative I've been
avoiding is to work this out by hand using the delta
method.

Here's my current code:

xtprobit poor hsize hsize2 television rechead roompp
if survey, i(clusterid)

* Predict probability of being poor and standard error
predictnl phat = predict(pu0), variance(varp)

* Calculate mean probability and mean variance by
target population, ignoring covariances
collapse (mean) phat varp, by(targetpop)

Thanks!
Gabriel

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index