INVESTIGADORES
MONTES ROJAS Gabriel Victorio
congresos y reuniones científicas
Título:
Modeling endogeneity bias using observables
Autor/es:
MONTES ROJAS, GABRIEL VICTORIO; GALVAO, ANTONIO; SONG, SUYONG
Lugar:
Victoria
Reunión:
Conferencia; First Conference in Econometric Theory at Universidad de San Andrés; 2013
Institución organizadora:
Center for Econometrics at Northwestern University and the Department of Economics at UdeSA
Resumen:
This paper develops practical inference methods for econometric models with endogeneity while relaxing the standard exclusion restrictions. We propose an alternative solution to the endogeneity problem by explicitly modeling the joint interaction of the endogenous variables and the unobserved causes of the dependent variable as a function of additional observables. These additional variables are defined as simultaneous variables because they are allowed to still be correlated with the innovation term. We derive identification of the parameters, develop an estimator, and establish its consistency and asymptotic normality. Inference procedures are based on the Wald statistic. We illustrate the methods with an example that examines the investment equation model. Our empirical findings support the idea that the proposed estimator is a useful alternative to the instrumental variables for economic applications in which endogeneity is a relevant concern.