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

st: negative binomial postest Q - how to predict X's setting Y=0?

From   "Ada Ma" <>
Subject   st: negative binomial postest Q - how to predict X's setting Y=0?
Date   Mon, 22 Sep 2008 01:43:42 +0100

Dear Statalisters,

I have estimated a negative binomial model using STATA's nbreg (I'm
using negbin2).  The model features a dependent variable Y, and a
bunch of explanatory variables which includes a policy explanatory
variable X.

X is continuous and can be positive or negative.  I want to calculate
the values of X's where Y would equal to zero.  Are there examples out
there that I can follow?  Are there potential pitfalls I might face?

My plan - daft but I imagine it should ensure that I avoid most
pitfalls - is to keep on increasing / decreasing X's by 1 and use the
revised X's to predict Y_hat.  If I try all the possible values (min X
to max X) then I should have a large number of predict Y_hats based on
the altered X's, if the predicted Y_hat is close to zero than the
corresponding altered X's should be the ones I'm shooting for.  I'd
love to be given suggestions on what are the smarter methods.

Thank a lot in advance!!


Ada Ma
Research Fellow
Health Economics Research Unit
University of Aberdeen, UK.
Tel: +44 (0) 1224 553863
Fax: +44 (0) 1224 550926
*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index