Efficient estimation with missing data and endogeneity. (7th February 2023)

• Record Type: Journal Article
• Authors: Rai, Bhavna
• Is Part Of: Econometric reviews. Volume 42:Number 2(2023)
• Page Start: 220
• Page End: 239
• Publication Date: 2023-02-07
• Subjects: Missing data -- endogeneity -- instrumental variables -- imputation -- nonlinearity; C13 -- C36; Econometrics -- Periodicals; 330.015195
• DOI: 10.1080/07474938.2023.2178089
• Languages: English
• ISSNs: 0747-4938

Abstract

Abstract: I study the problem of missing values in the outcome and endogenous covariates in linear models. I propose an estimator that improves efficiency relative to a complete cases 2SLS. Unlike traditional imputation, my estimator is consistent even if the model contains nonlinear functions – like squares and interactions – of the endogenous covariates. It can also be used to combine data sets with missing outcome, missing endogenous covariates, and no missing variables. It includes the well-known "Two-Sample 2SLS" as a special case under weaker assumptions than the corresponding literature.