High-dimensional instrumental variables regression and confidence sets

Abstract

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.

References in corpus (3)

• Square-Root Lasso: Pivotal Recovery of Sparse Signals via Conic Programming
• The Dantzig selector and sparsity oracle inequalities
• High Dimensional Sparse Econometric Models: An Introduction

Cited by in corpus (32)

• Challenges of Big Data Analysis
• Gaussian approximations and multiplier bootstrap for maxima of sums of high-dimensional random vectors
• On the Prediction Performance of the Lasso
• Valid Post-Selection and Post-Regularization Inference: An Elementary, General Approach
• Distributions of Angles in Random Packing on Spheres
• Pivotal estimation via square-root Lasso in nonparametric regression
• Endogeneity in high dimensions
• Regularization Methods for High-Dimensional Instrumental Variables Regression With an Application to Genetical Genomics
• Double/Debiased Machine Learning for Treatment and Causal Parameters
• A new scope of penalized empirical likelihood with high-dimensional estimating equations
• A General Theory of Hypothesis Tests and Confidence Regions for Sparse High Dimensional Models
• Minimax Estimation of Conditional Moment Models
• Identifying Network Ties from Panel Data: Theory and an Application to Tax Competition
• An autocovariance-based learning framework for high-dimensional functional time series
• Uniform Inference for High-dimensional Quantile Regression: Linear Functionals and Regression Rank Scores
• Pivotal Estimation via Self-Normalization for High-Dimensional Linear Models with Error in Variables
• Confidence Intervals for Causal Effects with Invalid Instruments using Two-Stage Hard Thresholding with Voting
• Learning Heteroscedastic Models by Convex Programming under Group Sparsity
• Minimum Distance Approach to Inference with Many Instruments
• High Dimensional Linear GMM
• High dimensional errors-in-variables models with dependent measurements
• Variable Selection with Exponential Weights and $l_0$-Penalization
• Adaptive Estimation in Structured Factor Models with Applications to Overlapping Clustering
• An $\{l_1,l_2,l_{\infty }\}$-Regularization Approach to High-Dimensional Errors-in-variables Models
• Pivotal estimation in high-dimensional regression via linear programming
• Detection and estimation of parameters in high dimensional multiple change point regression models via ${\ell }_1/{\ell }_0$ regularization and discrete optimization
• The $l_q$ consistency of the Dantzig Selector for Cox's Proportional Hazards Model
• Testing Endogeneity with High Dimensional Covariates
• Estimation of Discrete Choice Models: A Machine Learning Approach
• Causality and Generalizability: Identifiability and Learning Methods
• Inference robust to outliers with l1-norm penalization
• Errors-in-variables models with dependent measurements