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

RE: st: problem with the inteff command

From   "Nick Cox" <>
To   <>
Subject   RE: st: problem with the inteff command
Date   Wed, 5 Mar 2008 19:56:04 -0000

Given that the Statalist response to this question has now executed a perfect circle, 
you may be best advised to approach the program authors directly. 

Jochen Späth

once again thanks for your help. Unfortunately your suggestion did not help to solve the problem since the homepage you posted is exactly the site from which the ado-file was actually obtained-just for your information.

Maarten buis

I vaguely remember that that problem got solved when the latest version
of -inteff- was installed, which can be downloaded from Edward Norton's

Jochen Späth

> I try to estimate the interaction effect between two variables in a
> logit model. 
> My model looks like this: Pr(y) = x1 + x2 + x1*x2 + X
> x1 is a continuously variable, x2 is a dummy variable in some
> specifications, in others it is a continuously variable. x1*x2 is the
> interaction effect, X are several control variables.
> To obtain the correct interaction effect, I use the inteff command
> provided as an ado-file by Norton, Wang, and Ai (2004) "Computing
> interaction effects and standard errors in logit and probit models",
> The Stata Journal 4, Number 2, pp. 154-167.
> Unfortunately, in some specifications the inteff command doesn't work
> and displays following error message:
> conformability error
> r(503);
> This problem was on statalist in April 2006, but as far as I see
> there was no solution provided. (see
> Does
> anybody have an idea how to solve my problem?

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index