Smoothed GMM for quantile models

We consider estimation of finite-dimensional parameters identified by general conditional quantile restrictions, including instrumental variables quantile regression. Within a generalized method of moments framework, moment functions are smoothed to aid both computation and precision. Consistency an...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of econometrics 2019-11, Vol.213 (1), p.121-144
Hauptverfasser: de Castro, Luciano, Galvao, Antonio F., Kaplan, David M., Liu, Xin
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We consider estimation of finite-dimensional parameters identified by general conditional quantile restrictions, including instrumental variables quantile regression. Within a generalized method of moments framework, moment functions are smoothed to aid both computation and precision. Consistency and asymptotic normality are established under weaker assumptions than previously seen in the literature, allowing dependent data and nonlinear structural models. Simulations illustrate the finite-sample properties. An in-depth empirical application estimates the consumption Euler equation derived from quantile utility maximization. Advantages of quantile Euler equations include robustness to fat tails, decoupling risk attitude from the elasticity of intertemporal substitution, and error-free log-linearization.
ISSN:0304-4076
1872-6895
DOI:10.1016/j.jeconom.2019.04.008