Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/86534 
Authors: 
Year of Publication: 
2005
Series/Report no.: 
Tinbergen Institute Discussion Paper No. 05-084/4
Publisher: 
Tinbergen Institute, Amsterdam and Rotterdam
Abstract: 
In binary discrete regression models like logit or probit the omis-sion of a relevant regressor (even if it is orthogonal) depresses the re-maining b coefficients towards zero. For the probit model, Wooldridge(2002) has shown that this bias does not carry over to the effect ofthe regressor on the outcome. We find by simulations that this alsoholds for logit models, even when the omitted variable leads to severemisspecification of the disturbance. More simulations show that es-timates of these effects by logit analysis are also impervious to puremisspecification of the disturbance.
Subjects: 
logit model
omitted variables
misspecification
JEL: 
C25
Document Type: 
Working Paper

Files in This Item:
File
Size
215.33 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.