There is a typo in my edition of the book. I wonder whether that typo is the
source of your problem. The typo is that the variable names of the predict
statement is different from the variable name of the commands that follow.
If you have Stata 8, the following command work as in picture at page 172
use http://www.stata-press.com/data/lfr/ordwarm2.dta, clear
ologit warm yr89 male white age ed prst, nolog
predict SDwarm Dwarm Awarm SAwarm
label var SDwarm "Pr(SD)"
label var Dwarm "Pr(D)"
label var Awarm "Pr(A)"
label var SAwarm "Pr(SA)"
dotplot SDwarm Dwarm Awarm SAwarm, ylabel(0(.25).75) ysize(2.0124)
xsize(3.039)
The other possible cause of your issue is that you are using the version of
the book written for Stata 7 and using Stata 8.
If so you can continue to use the commands as they are written in your book,
but you must put -version 7- at the beginning or your do file (or -version
7:- before the commands that do not work).
If this does not help you, please post the exact commands you use to
generate the graphs and the first two lines of the output of the -about-
command.
Best,
Renzo Comolli
----------------------------------------------------------------------------
----
*From Phuong Nguyen <[email protected]>
To [email protected]
Subject st: Ploting Predicted Probability for Ologit
Date Fri, 26 Nov 2004 17:11:28 -0800
----------------------------------------------------------------------------
----
Hi there,
I wonder if you know the way to plot predicted probablity for ologit
regression? I followed chapter 5 in Long's book (Regression models for
categorical and limited dependent variables). As for now, I am able to
calculate all marginal probabilities at min, max, and average values
but I can't produce predicted and cummulative probabilities graphs as
in Figure 5.3.
thanks ,
Phuong
*
* 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/
