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

st: discriminatory power of a probit model

From   "Martin Weiss" <>
To   <>
Subject   st: discriminatory power of a probit model
Date   Sun, 2 Mar 2008 18:19:19 +0100

Dear Statalisters,

try this in Stata:

sysuse auto, clear
probit foreign weight, nolog
estat clas
probit foreign weight mpg, nolog
estat clas

Then tell me: can I prefer one specification of covariates in a probit model
over the other on the basis of the correctly classified cases as provided at
the bottom of the classification table? If so, is there a confidence
interval that would let me judge whether the difference between two models
is significant? And what is the difference to 

a) the area under the ROC curve, as provided by the -lroc- command ( and
where I do know that you can test two models against each other based on the
-roccomp- command)

b) the gofs provided by the -estat gof- command?

We recently heard a presentation at our workshop where someone argued in
this vein, and all I could come with was this lousy "I have a bad feeling
about this". 

Martin Weiss

*   For searches and help try:

© Copyright 1996–2019 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index