Ill-posed estimation in high-dimensional models with instrumental variables

DOI10.1016/J.JECONOM.2020.04.043zbMATH Open1464.62346arXiv1806.00666OpenAlexW3048281013MaRDI QIDQ2227078FDOQ2227078

Authors: Christoph Breunig, Enno Mammen, Anna Simoni 

Publication date: 9 February 2021

Published in: Journal of Econometrics (Search for Journal in Brave)

Abstract: This paper is concerned with inference about low-dimensional components of a high-dimensional parameter vector $$ which is identified through instrumental variables. We allow for eigenvalues of the expected outer product of included and excluded covariates, denoted by $M$, to shrink to zero as the sample size increases. We propose a novel estimator based on desparsification of an instrumental variable Lasso estimator, which is a regularized version of 2SLS with an additional correction term. This estimator converges to $$ at a rate depending on the mapping properties of $M$ captured by a sparse link condition. Linear combinations of our estimator of $$ are shown to be asymptotically normally distributed. Based on consistent covariance estimation, our method allows for constructing confidence intervals and statistical tests for single or low-dimensional components of $$. In Monte-Carlo simulations we analyze the finite sample behavior of our estimator.

Full work available at URL: https://arxiv.org/abs/1806.00666

Recommendations

• Inference for high-dimensional instrumental variables regression
• Instrumental variable estimation of nonseparable models
• On ill-posedness of nonparametric instrumental variable regression with convexity constraints
• Efficient Instrumental Variables Estimation of Nonlinear Models
• Inference in instrumental variable models with heteroskedasticity and many instruments
• Instrumental variables estimation and inference in the presence of many exogenous regressors
• On parameter estimation for high dimensional errors-in-variables models
• Instrumental variable estimation with heteroskedasticity and many instruments
• High-dimensional inference in misspecified linear models

zbMATH Keywords

Lassolinear modelcentral limit theoreminstrumental variablessparsitydesparsificationill-posed estimation problem

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Applications of statistics to economics (62P20) Asymptotic distribution theory in statistics (62E20) Ridge regression; shrinkage estimators (Lasso) (62J07)

Cites Work

• Statistics for high-dimensional data. Methods, theory and applications.
• High-dimensional graphs and variable selection with the Lasso
• Confidence Intervals and Hypothesis Testing for High-Dimensional Regression
• Double/debiased machine learning for treatment and structural parameters
• Confidence Intervals for Low Dimensional Parameters in High Dimensional Linear Models
• On asymptotically optimal confidence regions and tests for high-dimensional models
• Covariance regularization by thresholding
• Endogeneity in high dimensions
• Estimation and testing under sparsity. École d'Été de Probabilités de Saint-Flour XLV -- 2015
• Sparse models and methods for optimal instruments with an application to eminent domain
• Hypothesis Testing in High-Dimensional Regression Under the Gaussian Random Design Model: Asymptotic Theory
• Scaled sparse linear regression
• Statistical significance in high-dimensional linear models
• Program evaluation and causal inference with high-dimensional data
• Consistent Estimation with a Large Number of Weak Instruments
• Efficient Estimation of Models with Conditional Moment Restrictions Containing Unknown Functions
• Confidence intervals for causal effects with invalid instruments by using two-stage hard thresholding with voting
• Optimal sup-norm rates and uniform inference on nonlinear functionals of nonparametric IV regression
• Nonparametric instrumental regression
• Instrumental Variable Estimation of Nonparametric Models
• Minimax estimation of large covariance matrices under \(\ell_1\)-norm
• Adaptive estimation of functionals in nonparametric instrumental regression
• Automobile Prices in Market Equilibrium
• Semi-Nonparametric IV Estimation of Shape-Invariant Engel Curves
• Instrumental variables estimation with many weak instruments using regularized JIVE
• Estimation and inference with weak, semi-strong, and strong identification
• Estimation of nonparametric conditional moment models with possibly nonsmooth generalized residuals
• On the use of the Lasso for instrumental variables estimation with some invalid instruments
• Selecting instrumental variables in a data rich environment
• A unified theory of confidence regions and testing for high-dimensional estimating equations
• Sieve Wald and QLR inferences on semi/nonparametric conditional moment models

Cited In (7)

• Endogenous treatment effect estimation using high-dimensional instruments and double selection
• Inference for high-dimensional instrumental variables regression
• Regularization methods for high-dimensional instrumental variables regression with an application to genetical genomics
• Estimating the effect of a variable in a high-dimensional linear model
• Sparse linear models and \(l_1\)-regularized 2SLS with high-dimensional endogenous regressors and instruments
• Testing endogeneity with high dimensional covariates
• On the use of the Lasso for instrumental variables estimation with some invalid instruments

This page was built for publication: Ill-posed estimation in high-dimensional models with instrumental variables

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2227078)