Approximating the marginal effect of discrete regressors in logit models

Vince Daly

Research output: Working paperDiscussion paper

Abstract

Logit models are non-linear in their explanatory variables. Derivatives with respect to the explanatory variables therefore only approximate the response to discrete changes in regressor values, yet have gained some support within the literature. This note investigates such an approximation strategy and delineates the conditions under which it does or does not give satisfactory results. The conclusion is that it is a risky strategy and that therefore the correct alternative should be employed.
Original languageEnglish
Place of PublicationKingston upon Thames, U.K.
PublisherFaculty of Arts and Social Sciences, Kingston University
Number of pages21
Publication statusPublished - Aug 2006
Externally publishedYes

Publication series

NameEconomics Discussion Paper
PublisherFaculty of Arts and Social Sciences, Kingston University

Bibliographical note

Note: Economics discussion paper, 2006/5

Keywords

  • Economics and econometrics
  • approximation error

Fingerprint

Dive into the research topics of 'Approximating the marginal effect of discrete regressors in logit models'. Together they form a unique fingerprint.

Cite this