Using invalid instruments on purpose: focused moment selection and averaging for GMM

DOI10.1016/J.JECONOM.2016.07.006zbMATH Open1443.62441arXiv1408.0705OpenAlexW3124377371MaRDI QIDQ337769FDOQ337769

Francis J. DiTraglia

Publication date: 3 November 2016

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

Abstract: In finite samples, the use of a slightly endogenous but highly relevant instrument can reduce mean-squared error (MSE). Building on this observation, I propose a novel moment selection procedure for GMM -- the Focused Moment Selection Criterion (FMSC) -- in which moment conditions are chosen not based on their validity but on the MSE of their associated estimator of a user-specified target parameter. The FMSC mimics the situation faced by an applied researcher who begins with a set of relatively mild "baseline" assumptions and must decide whether to impose any of a collection of stronger but more controversial "suspect" assumptions. When the (correctly specified) baseline moment conditions identify the model, the FMSC provides an asymptotically unbiased estimator of asymptotic MSE, allowing us to select over the suspect moment conditions. I go on to show how the framework used to derive the FMSC can address the problem of inference post-moment selection. Treating post-selection estimators as a special case of moment-averaging, in which estimators based on different moment sets are given data-dependent weights, I propose simulation-based procedures for inference that can be applied to a variety of formal and informal moment-selection and averaging procedures. Both the FMSC and confidence interval procedures perform well in simulations. I conclude with an empirical example examining the effect of instrument selection on the estimated relationship between malaria and income per capita.

Full work available at URL: https://arxiv.org/abs/1408.0705

Recommendations

• Detecting invalid instruments using \(L_{1}\)-GMM
• Efficient GMM with nearly-weak instruments
• Instrumental variable and GMM estimation for panel data with measurement error
• Factor-GMM estimation with large sets of possibly weak instruments
• Estimating structural mean models with multiple instrumental variables using the generalised method of moments
• Salvaging falsified instrumental variable models
• On the use of the Lasso for instrumental variables estimation with some invalid instruments
• Improved instrumental variables and generalized method of moments estimators
• Instrumental variables methods with heterogeneity and mismeasured instruments

zbMATH Keywords

model averagingfocused information criterionGMM estimationmoment selectionpost-selection estimators

Mathematics Subject Classification ID

Applications of statistics to economics (62P20)

Cites Work

• Choosing the Number of Instruments
• P Values Maximized Over a Confidence Set for the Nuisance Parameter
• Generalized method of moments specification testing
• Sparse estimators and the oracle property, or the return of Hodges' estimator
• Can the strengths of AIC and BIC be shared? A conflict between model indentification and regression estimation
• THE IMPACT OF A HAUSMAN PRETEST ON THE ASYMPTOTIC SIZE OF A HYPOTHESIS TEST
• A Consistent Method for the Selection of Relevant Instruments
• ON THE ASYMPTOTIC SIZE DISTORTION OF TESTS WHEN INSTRUMENTS LOCALLY VIOLATE THE EXOGENEITY ASSUMPTION
• MODEL SELECTION AND INFERENCE: FACTS AND FICTION
• Model Selection and Model Averaging
• Frequentist Model Average Estimators
• The Focused Information Criterion
• On the Large-Sample Minimal Coverage Probability of Confidence Intervals After Model Selection
• VAR forecasting under misspecification
• Consistent Moment Selection Procedures for Generalized Method of Moments Estimation
• Consistent model and moment selection procedures for GMM estimation with application to dynamic panel data models
• Estimation and inference in the case of competing sets of estimating equations
• The Existence of Moments of k-Class Estimators
• The Exact Distribution of Instrumental Variable Estimators in an Equation Containing n + 1 Endogenous Variables
• GENERALIZED EMPIRICAL LIKELIHOOD–BASED MODEL SELECTION CRITERIA FOR MOMENT CONDITION MODELS
• LASSO-TYPE GMM ESTIMATOR
• ADAPTIVE GMM SHRINKAGE ESTIMATION WITH CONSISTENT MOMENT SELECTION
• Are ``nearly exogenous instruments reliable?
• Near exogeneity and weak identification in generalized empirical likelihood estimators: many moment asymptotics
• Choosing instrumental variables in conditional moment restriction models
• Constructing Optimal Instruments by First-Stage Prediction Averaging
• MINIMIZING AVERAGE RISK IN REGRESSION MODELS
• Model Selection
• A New Method for Testing Separate Families of Hypotheses
• A Test in the Presence of Nuisance Parameters
• Variable Selection for Logistic Regression Using a Prediction‐Focused Information Criterion
• Stein-like 2SLS estimator
• Institutions rule: the primacy of institutions over geography and integration in economic development
• The validity of instruments revisited
• The weighted method of moments approach for moment condition models

Cited In (23)

• Model averaging, asymptotic risk, and regressor groups
• Focused information criterion for locally misspecified vector autoregressive models
• Model averaging in semiparametric estimation of treatment effects
• A weighted average limited information maximum likelihood estimator
• Confidence intervals for intentionally biased estimators
• Focused Information Criterion and Model Averaging for Large Panels With a Multifactor Error Structure
• Asymptotic refinements of a misspecification-robust bootstrap for generalized method of moments estimators
• Simultaneous estimation and inference for multiple response variables
• Focused estimation for noisy and small data sets: a Bayesian minimum expected loss estimator approach
• The robust focused information criterion for strong mixing stochastic processes with \(\mathscr{L}^2\)-differentiable parametric densities
• Efficient shrinkage in parametric models
• A model‐averaging treatment of multiple instruments in Poisson models with errors
• Over-identified doubly robust identification and estimation
• Culling the Herd of Moments with Penalized Empirical Likelihood
• A Consistent Variance Estimator for 2SLS When Instruments Identify Different LATEs
• Valid instrumental variable selection method using negative control outcomes and constructing efficient estimator
• Joint inference based on Stein-type averaging estimators in the linear regression model
• Distribution theory of the least squares averaging estimator
• Select the valid and relevant moments: an information-based Lasso for GMM with many moments
• Model averaging based on generalized method of moments
• INFERENCE AFTER MODEL AVERAGING IN LINEAR REGRESSION MODELS
• An averaging estimator for two-step m-estimation in semiparametric models
• Optimal Model Averaging Based on Generalized Method of Moments

This page was built for publication: Using invalid instruments on purpose: focused moment selection and averaging for GMM

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q337769)