A Practical Comparison of the Bivariate Probit and Linear IV Estimators

This paper presents asymptotic theory and Monte-Carlo simulations comparing maximum-likelihood bivariate probit and linear instrumental variables estimators of treatment effects in models with a binary endogenous treatment and binary outcome. The t...

Full description

Bibliographic Details
Main Authors: Chiburis, Richard C., Das, Jishnu, Lokshin, Michael
Format: Policy Research Working Paper
Language:English
Published: 2012
Subjects:
Online Access:http://www-wds.worldbank.org/external/default/main?menuPK=64187510&pagePK=64193027&piPK=64187937&theSitePK=523679&menuPK=64187510&searchMenuPK=64187283&siteName=WDS&entityID=000158349_20110317174628
http://hdl.handle.net/10986/3368
id okr-10986-3368
recordtype oai_dc
spelling okr-10986-33682021-04-23T14:02:09Z A Practical Comparison of the Bivariate Probit and Linear IV Estimators Chiburis, Richard C. Das, Jishnu Lokshin, Michael ASYMPTOTIC DISTRIBUTION BOOTSTRAP CONFIDENCE INTERVALS CRITICAL VALUE CRITICAL VALUES DEGREES OF FREEDOM DEVELOPMENT RESEARCH DISTRIBUTION FUNCTION DUMMY VARIABLE ECONOMETRICS ENDOGENOUS REGRESSORS EQUATION SYSTEM ERROR ERROR TERMS ESTIMATORS EXHIBITS GOODNESS OF FIT HYPOTHESIS TESTING INSTRUMENTAL VARIABLES INSTRUMENTAL VARIABLES ESTIMATION KURTOSIS LIMITED DEPENDENT VARIABLE LIMITED DEPENDENT VARIABLES LOG-LIKELIHOOD FUNCTION LOGISTIC REGRESSION MATRIX MAXIMUM LIKELIHOOD MEAN SQUARE 0 HYPOTHESIS NUMBER OF OBSERVATIONS PREDICTION PROBABILITIES PROBABILITY PUBLIC SERVICES RANDOM VARIABLES REGRESSION MODEL RESEARCH WORKING PAPERS RESEARCHERS SAMPLE SIZE SCIENCES SIMULATION SIMULATIONS SKEWNESS SMALL SAMPLE STANDARD DEVIATION STANDARD ERRORS STANDARD NORMAL DISTRIBUTION STATA STATISTICAL SOFTWARE STRUCTURAL PARAMETERS T-TESTS TECHNIQUES TEST STATISTIC VALIDITY This paper presents asymptotic theory and Monte-Carlo simulations comparing maximum-likelihood bivariate probit and linear instrumental variables estimators of treatment effects in models with a binary endogenous treatment and binary outcome. The three main contributions of the paper are (a) clarifying the relationship between the Average Treatment Effect obtained in the bivariate probit model and the Local Average Treatment Effect estimated through linear IV; (b) comparing the mean-square error and the actual size and power of tests based on these estimators across a wide range of parameter values relative to the existing literature; and (c) assessing the performance of misspecification tests for bivariate probit models. The authors recommend two changes to common practices: bootstrapped confidence intervals for both estimators, and a score test to check goodness of fit for the bivariate probit model. 2012-03-19T18:01:11Z 2012-03-19T18:01:11Z 2011-03-01 http://www-wds.worldbank.org/external/default/main?menuPK=64187510&pagePK=64193027&piPK=64187937&theSitePK=523679&menuPK=64187510&searchMenuPK=64187283&siteName=WDS&entityID=000158349_20110317174628 http://hdl.handle.net/10986/3368 English Policy Research working paper ; no. WPS 5601 CC BY 3.0 IGO http://creativecommons.org/licenses/by/3.0/igo/ World Bank Publications & Research :: Policy Research Working Paper The World Region The World Region
repository_type Digital Repository
institution_category Foreign Institution
institution Digital Repositories
building World Bank Open Knowledge Repository
collection World Bank
language English
topic ASYMPTOTIC DISTRIBUTION
BOOTSTRAP
CONFIDENCE INTERVALS
CRITICAL VALUE
CRITICAL VALUES
DEGREES OF FREEDOM
DEVELOPMENT RESEARCH
DISTRIBUTION FUNCTION
DUMMY VARIABLE
ECONOMETRICS
ENDOGENOUS REGRESSORS
EQUATION SYSTEM
ERROR
ERROR TERMS
ESTIMATORS
EXHIBITS
GOODNESS OF FIT
HYPOTHESIS TESTING
INSTRUMENTAL VARIABLES
INSTRUMENTAL VARIABLES ESTIMATION
KURTOSIS
LIMITED DEPENDENT VARIABLE
LIMITED DEPENDENT VARIABLES
LOG-LIKELIHOOD FUNCTION
LOGISTIC REGRESSION
MATRIX
MAXIMUM LIKELIHOOD
MEAN SQUARE
0 HYPOTHESIS
NUMBER OF OBSERVATIONS
PREDICTION
PROBABILITIES
PROBABILITY
PUBLIC SERVICES
RANDOM VARIABLES
REGRESSION MODEL
RESEARCH WORKING PAPERS
RESEARCHERS
SAMPLE SIZE
SCIENCES
SIMULATION
SIMULATIONS
SKEWNESS
SMALL SAMPLE
STANDARD DEVIATION
STANDARD ERRORS
STANDARD NORMAL DISTRIBUTION
STATA
STATISTICAL SOFTWARE
STRUCTURAL PARAMETERS
T-TESTS
TECHNIQUES
TEST STATISTIC
VALIDITY
spellingShingle ASYMPTOTIC DISTRIBUTION
BOOTSTRAP
CONFIDENCE INTERVALS
CRITICAL VALUE
CRITICAL VALUES
DEGREES OF FREEDOM
DEVELOPMENT RESEARCH
DISTRIBUTION FUNCTION
DUMMY VARIABLE
ECONOMETRICS
ENDOGENOUS REGRESSORS
EQUATION SYSTEM
ERROR
ERROR TERMS
ESTIMATORS
EXHIBITS
GOODNESS OF FIT
HYPOTHESIS TESTING
INSTRUMENTAL VARIABLES
INSTRUMENTAL VARIABLES ESTIMATION
KURTOSIS
LIMITED DEPENDENT VARIABLE
LIMITED DEPENDENT VARIABLES
LOG-LIKELIHOOD FUNCTION
LOGISTIC REGRESSION
MATRIX
MAXIMUM LIKELIHOOD
MEAN SQUARE
0 HYPOTHESIS
NUMBER OF OBSERVATIONS
PREDICTION
PROBABILITIES
PROBABILITY
PUBLIC SERVICES
RANDOM VARIABLES
REGRESSION MODEL
RESEARCH WORKING PAPERS
RESEARCHERS
SAMPLE SIZE
SCIENCES
SIMULATION
SIMULATIONS
SKEWNESS
SMALL SAMPLE
STANDARD DEVIATION
STANDARD ERRORS
STANDARD NORMAL DISTRIBUTION
STATA
STATISTICAL SOFTWARE
STRUCTURAL PARAMETERS
T-TESTS
TECHNIQUES
TEST STATISTIC
VALIDITY
Chiburis, Richard C.
Das, Jishnu
Lokshin, Michael
A Practical Comparison of the Bivariate Probit and Linear IV Estimators
geographic_facet The World Region
The World Region
relation Policy Research working paper ; no. WPS 5601
description This paper presents asymptotic theory and Monte-Carlo simulations comparing maximum-likelihood bivariate probit and linear instrumental variables estimators of treatment effects in models with a binary endogenous treatment and binary outcome. The three main contributions of the paper are (a) clarifying the relationship between the Average Treatment Effect obtained in the bivariate probit model and the Local Average Treatment Effect estimated through linear IV; (b) comparing the mean-square error and the actual size and power of tests based on these estimators across a wide range of parameter values relative to the existing literature; and (c) assessing the performance of misspecification tests for bivariate probit models. The authors recommend two changes to common practices: bootstrapped confidence intervals for both estimators, and a score test to check goodness of fit for the bivariate probit model.
format Publications & Research :: Policy Research Working Paper
author Chiburis, Richard C.
Das, Jishnu
Lokshin, Michael
author_facet Chiburis, Richard C.
Das, Jishnu
Lokshin, Michael
author_sort Chiburis, Richard C.
title A Practical Comparison of the Bivariate Probit and Linear IV Estimators
title_short A Practical Comparison of the Bivariate Probit and Linear IV Estimators
title_full A Practical Comparison of the Bivariate Probit and Linear IV Estimators
title_fullStr A Practical Comparison of the Bivariate Probit and Linear IV Estimators
title_full_unstemmed A Practical Comparison of the Bivariate Probit and Linear IV Estimators
title_sort practical comparison of the bivariate probit and linear iv estimators
publishDate 2012
url http://www-wds.worldbank.org/external/default/main?menuPK=64187510&pagePK=64193027&piPK=64187937&theSitePK=523679&menuPK=64187510&searchMenuPK=64187283&siteName=WDS&entityID=000158349_20110317174628
http://hdl.handle.net/10986/3368
_version_ 1764386869375664128