Exploratory Analysis of Endogeneity in Discrete Choice Models
A model is said to be affected by endogeneity when its deterministic part is correlated with the error term. This is an issue that affects both linear models such as regression and non-linear models like discrete choice models. It is a classical and well studied problem in regression models, in particular in econometric literature. However, how to deal with endogeneity in discrete choice models is not as developed, and in particular still less has been done regarding endogeneity and transportation. The aim of this paper is to review what has already been developed in literature in order to identify potential contributions in this field. The paper contains a literature review and a description of three of the most common developed methods to deal with endogeneity.
Record created on 2014-10-23, modified on 2017-02-16