Non/Semi-parametric Instrumental Variables Models for Economic Data

In recent years, econometricians extended
instrumental variable models, a standard approach to
regression with endogeneity, to non/semi-parametric
settings. However, when the endogenous variables are
continuous, such models are exposed to ill-posed
problems. Working from a different assumption, this
book considers estimation of non/semi-parametric
instrumental variables models under a functional
coefficient form. Under this representation, models
are linear in the endogenous components with either
constant or unknown functional coefficients. Due to
the linearity, the models avoid the ill-posed
problems and at the same time retain the flexibility
of the regression function. This study constructs
the constant and functional coefficients estimators
and proves their consistency and asymptotic
normality. The high practical power of these
estimators is illustrated via Monte Carlo
simulations and an application to labor statistics.
The book provides a more efficient way for
researchers and practitioners to analyze economic
data with endogenous variables utilizing non/semi-
parametric models.

Autorentext
Huaiyu Xiong received a Bachelor of Science from Nankai
University, Tianjin, China, where he studied mathematics at the
Chern Institute of Mathematics. He also holds a Master of Arts
in Mathematics and Doctor of Philosophy with a concentration in
statistics from the University of North Carolina at Charlotte.

Klappentext
In recent years, econometricians extended instrumental variable models, a standard approach to regression with endogeneity, to non/semi-parametric settings. However, when the endogenous variables are continuous, such models are exposed to ill-posed problems. Working from a different assumption, this book considers estimation of non/semi-parametric instrumental variables models under a functional coefficient form. Under this representation, models are linear in the endogenous components with either constant or unknown functional coefficients. Due to the linearity, the models avoid the ill-posed problems and at the same time retain the flexibility of the regression function. This study constructs the constant and functional coefficients estimators and proves their consistency and asymptotic normality. The high practical power of these estimators is illustrated via Monte Carlo simulations and an application to labor statistics. The book provides a more efficient way for researchers and practitioners to analyze economic data with endogenous variables utilizing non/semi- parametric models.