Assessing Sensitivity to Unconfoundedness: Estimation and Inference

This paper provides a set of methods for quantifying the robustness of treatment effects estimated using the unconfoundedness assumption (also known as selection on observables or conditional independence). Specifically, we estimate and do inference on bounds on various treatment effect parameters, like the average treatment effect (ATE) and the average effect of treatment on the treated (ATT), under nonparametric relaxations of the unconfoundedness assumption indexed by a scalar sensitivity parameter c. These relaxations allow for limited selection on unobservables, depending on the value of c. For large enough c, these bounds equal the no assumptions bounds. Using a non-standard bootstrap method, we show how to construct confidence bands for these bound functions which are uniform over all values of c. We illustrate these methods with an empirical application to effects of the National Supported Work Demonstration program. We implement these methods in a companion Stata module for easy use in practice. Classification- C14, C18, C21, C51

[1]  Han Hong,et al.  The Numerical Delta Method , 2018, Journal of Econometrics.

[2]  G. W. Imbens.  Sensitivity to Exogeneity Assumptions in Program Evaluation , 2003 .

[3]  A. Ichino,et al.  From Temporary Help Jobs to Permanent Employment: What Can We Learn from Matching Estimators and Their Sensitivity? , 2006, SSRN Electronic Journal.

[4]  C. Manski.  Nonparametric Bounds on Treatment Effects , 1989 .

[5]  M. Kosorok.  Introduction to Empirical Processes and Semiparametric Inference , 2008 .

[6]  R. Lalonde.  Evaluating the Econometric Evaluations of Training Programs with Experimental Data , 1984 .

[7]  R. Mauro.  Understanding L.O.V.E. (left out variables error): A method for estimating the effects of omitted variables. , 1990 .

[8]  Andrés Santos,et al.  Inference on Directionally Differentiable Functions , 2014, The Review of Economic Studies.

[9]  D. Rubin,et al.  Causal Inference for Statistics, Social, and Biomedical Sciences: Sensitivity Analysis and Bounds , 2015 .

[10]  D. Rubin,et al.  Assessing Sensitivity to an Unobserved Binary Covariate in an Observational Study with Binary Outcome , 1983 .

[11]  E. Oster.  Unobservable Selection and Coefficient Stability: Theory and Evidence , 2019 .

[12]  Carlos Cinelli,et al.  Making sense of sensitivity: extending omitted variable bias , 2019, Journal of the Royal Statistical Society: Series B (Statistical Methodology).

[13]  Brian Krauth,et al.  Bounding a Linear Causal Effect Using Relative Correlation Restrictions , 2015 .

[14]  Christopher R. Taber,et al.  Using Selection on Observed Variables to Assess Bias from Unobservables When Evaluating Swan-Ganz Catheterization , 2008 .

[15]  Xiaojie Mao,et al.  Interval Estimation of Individual-Level Causal Effects Under Unobserved Confounding , 2018, AISTATS.

[16]  Jeffrey A. Smith,et al.  Evaluating the Welfare State , 1998 .

[17]  Soumendu Sundar Mukherjee,et al.  Weak convergence and empirical processes , 2019 .

[18]  Alexandre Poirier,et al.  Identification of Treatment Effects under Conditional Partial Independence , 2017, 1707.09563.

[19]  J. Robins,et al.  Sensitivity Analysis for Selection bias and unmeasured Confounding in missing Data and Causal inference models , 2000 .

[20]  Christopher R. Taber,et al.  Selection on Observed and Unobserved Variables: Assessing the Effectiveness of Catholic Schools , 2000, Journal of Political Economy.

[21]  Paul W. Holland,et al.  The sensitivity of linear regression coefficients' confidence limits to the omission of a confounder , 2009, 0905.3463.

[22]  Whitney K. Newey,et al.  LARGE SAMPLE ESTIMATION AND HYPOTHESIS , 1999 .

[23]  J. Heckman,et al.  The Economics and Econometrics of Active Labor Market Programs , 1999 .