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

Re: st: How to make prediction after a spline regression

From   Roger Newson <>
Subject   Re: st: How to make prediction after a spline regression
Date   Tue, 13 Apr 2004 12:16:54 +0100

At 11:34 09/04/04 -0700, Shige wrote:
Dear All,

I would like to follow my previous question about how to use spline.
Suppose I create a spline using "frencurv" and then estimate a logistic
regression like this:

. frencurv, xvar(school) refpts(0(4)16) gen(sch) power(1)
. xtlogit unemp sch* X, re i(id) noconstant

How to generated predicted probablity so that I can create a two-way plot
of "prob(unemp) by school", like the one in pp.7 in bspline user guide?
Thank you very much!
Figure 3 in the -bspline- user guide was created using -predict- as follows (under Stata 6):

* frencurv with extended reference points (default) and cubic spline *
frencurv,xvar(weight) refpts(1760(770)4840) gen(cs) power(3)
describe cs*
regress mpg cs*,noconst robust
lincom cs3-cs5
predict mpghat3
grap mpg mpghat3 weight,s(o.) c(.L) xlab(1760(770)4840) ylab(0(5)45) t1(" ") l1("Mileage (mpg)") saving(fig3.gph,replace)

Therefore, use -predict- after -xtlogit- to get predicted probabilities in the usual way.


Roger Newson
Lecturer in Medical Statistics
Department of Public Health Sciences
King's College London
5th Floor, Capital House
42 Weston Street
London SE1 3QD
United Kingdom

Tel: 020 7848 6648 International +44 20 7848 6648
Fax: 020 7848 6620 International +44 20 7848 6620
or 020 7848 6605 International +44 20 7848 6605

Opinions expressed are those of the author, not the institution.

* For searches and help try:

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