Marginal Effects for Probit and Tobit with Endogeneity

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Evdokimov, Kirill S., Kalnina, Ilze, Zeleneev, Andrei
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912175658893312
author Evdokimov, Kirill S.
Kalnina, Ilze
Zeleneev, Andrei
author_facet Evdokimov, Kirill S.
Kalnina, Ilze
Zeleneev, Andrei
contents When evaluating partial effects, it is important to distinguish between structural endogeneity and measurement errors. In contrast to linear models, these two sources of endogeneity affect partial effects differently in nonlinear models. We study this issue focusing on the Instrumental Variable (IV) Probit and Tobit models. We show that even when a valid IV is available, failing to differentiate between the two types of endogeneity can lead to either under- or over-estimation of the partial effects. We develop simple estimators of the bounds on the partial effects and provide easy to implement confidence intervals that correctly account for both types of endogeneity. We illustrate the methods in a Monte Carlo simulation and an empirical application.
format Preprint
id arxiv_https___arxiv_org_abs_2306_14862
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Marginal Effects for Probit and Tobit with Endogeneity
Evdokimov, Kirill S.
Kalnina, Ilze
Zeleneev, Andrei
Econometrics
Methodology
When evaluating partial effects, it is important to distinguish between structural endogeneity and measurement errors. In contrast to linear models, these two sources of endogeneity affect partial effects differently in nonlinear models. We study this issue focusing on the Instrumental Variable (IV) Probit and Tobit models. We show that even when a valid IV is available, failing to differentiate between the two types of endogeneity can lead to either under- or over-estimation of the partial effects. We develop simple estimators of the bounds on the partial effects and provide easy to implement confidence intervals that correctly account for both types of endogeneity. We illustrate the methods in a Monte Carlo simulation and an empirical application.
title Marginal Effects for Probit and Tobit with Endogeneity
topic Econometrics
Methodology
url https://arxiv.org/abs/2306.14862