Instrumental variable estimation of nonlinear models with nonclassical measurement error using control variables

被引:6
|
作者
Hahn, Jinyong [1 ]
Ridder, Geert [2 ]
机构
[1] UCLA, Dept Econ, Bunche Hall, Los Angeles, CA 90095 USA
[2] Univ Southern Calif, Dept Econ, Kaprilian Hall, Los Angeles, CA 90089 USA
基金
美国国家科学基金会;
关键词
Measurement error; Endogenous regressor; Control variables; Semi-parametric estimation; Influence function; SEMIPARAMETRIC ESTIMATORS; ASYMPTOTIC VARIANCE; IDENTIFICATION; MISCLASSIFICATION; INFORMATION; REGRESSION;
D O I
10.1016/j.jeconom.2017.06.008
中图分类号
F [经济];
学科分类号
02 ;
摘要
We consider nonlinear parametric models with an independent variable that is measured with error. The measurement error can be correlated with the true value, i.e., the measurement error is allowed to be nonclassical. We propose a control variable estimator for the parameters of interest. The estimator is consistent even if the latent true value is endogenous. We derive the influence function of the semi parametric estimator that accounts for the estimation of the control variable in the first stage. (C) 2017 Elsevier B.V.All rights reserved.
引用
收藏
页码:238 / 250
页数:13
相关论文
共 50 条