High-dimensional instrumental variables regression and confidence sets

This article considers inference in linear instrumental variables models with many regressors, all of which could be endogenous. We propose the STIV estimator. Identification robust confidence sets are derived by solving linear programs. We present results on rates of convergence, variable selection...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2021-08
Main Authors: Gautier, Eric, Rose, Christiern
Format: Article
Language:English
Subjects:
Economic models
Error analysis
Error detection
Functionals
Regression analysis
Sensitivity analysis
Sparsity
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This article considers inference in linear instrumental variables models with many regressors, all of which could be endogenous. We propose the STIV estimator. Identification robust confidence sets are derived by solving linear programs. We present results on rates of convergence, variable selection, confidence sets which adapt to the sparsity, and analyze confidence bands for vectors of linear functions using bias correction. We also provide solutions to some instruments being endogenous. The application is to the EASI demand system.
ISSN:2331-8422