Abstract

http://ssrn.com/abstract=1668750
 
 

References (12)



 
 

Citations (7)



 


 



On the Economic Meaning of Interaction Term Coefficients in Non-Linear Binary Response Regression Models


Adam C. Kolasinski


Texas A&M, Mays School of Business

Andrew F. Siegel


University of Washington - Department of Finance and Business Economics; National Bureau of Economic Research (NBER)

August 30, 2010


Abstract:     
We show that it is perfectly correct to use just the interaction term, along with its standard error, to draw inferences about interactive effects in binary response regression models. This point is currently in dispute among applied econometricians, some of whom insist that simply relying on the interaction term is incorrect, since the cross partial derivative of the probability of occurrence with respect to interacted covariates can, for some observations, have the sign opposite to that of the interaction term coefficient. We show that this sign flip results from a mechanical saturation effect that is of no importance to researchers who recognize that small changes in probability are more important near the boundaries than near the center. For such researchers, the interaction term coefficient (which is the cross partial derivative of the logit or probit function of the probability) provides a more meaningful measure of interactive effects than does the cross partial derivative of the probability itself. We introduce an alternative cross partial derivative of the probability for which these sign changes cannot occur. Finally, we demonstrate some simple and intuitive ways of interpreting the economic meaning of interaction term coefficients.

Number of Pages in PDF File: 26

Keywords: Logit regresion, probit regression, interactive effects

JEL Classification: C25, C12

working papers series





Download This Paper

Date posted: August 31, 2010 ; Last revised: October 27, 2010

Suggested Citation

Kolasinski, Adam C. and Siegel, Andrew F., On the Economic Meaning of Interaction Term Coefficients in Non-Linear Binary Response Regression Models (August 30, 2010). Available at SSRN: http://ssrn.com/abstract=1668750 or http://dx.doi.org/10.2139/ssrn.1668750

Contact Information

Adam C. Kolasinski (Contact Author)
Texas A&M, Mays School of Business ( email )
360 Wehner
College Station, TX 77843-4218
United States
Andrew F. Siegel
University of Washington - Department of Finance and Business Economics ( email )
Box 353200
Seattle, WA 98195
United States
National Bureau of Economic Research (NBER)
1050 Massachusetts Avenue
Cambridge, MA 02138
United States
Feedback to SSRN


Paper statistics
Abstract Views: 2,307
Downloads: 403
Download Rank: 40,773
References:  12
Citations:  7

© 2014 Social Science Electronic Publishing, Inc. All Rights Reserved.  FAQ   Terms of Use   Privacy Policy   Copyright   Contact Us
This page was processed by apollo1 in 0.297 seconds