A Comparison of Inferential Techniques for Instrumental Variables Methods

In randomized experiments, subjects often fail to comply with the assigned treatment assignment. When such non-compliance occurs, the method of instrumental variables provides a framework to study causal effects for those who actually received the treatment. In this paper, we compare various finite sample methods of inference used in instrumental variables. We begin our comparison with an exact method, which uses the randomization in experiments as a basis for inference, but lacks a closed-form solution and may be computationally infeasible. We then provide alternatives to the exact method, including the almost exact method which is computationally feasible but retains the advantages of the exact method. We also discuss the most widespread method for inference using asymptotic Normal approximations and inference based on fixed compliance rate. We then compare all four methods using a set of simulations. We conclude with comparisons based on three different applications from the social sciences.

[1]  Dylan S. Small,et al.  Randomization‐based instrumental variables methods for binary outcomes with an application to the ‘IMPROVE’ trial , 2017 .

[2]  D. Rubin,et al.  Causal Inference for Statistics, Social, and Biomedical Sciences: An Introduction , 2016 .

[3]  Fan Yang,et al.  Dissonant Conclusions When Testing the Validity of an Instrumental Variable , 2014 .

[4]  Dylan S Small,et al.  Using an instrumental variable to test for unmeasured confounding , 2014, Statistics in medicine.

[5]  M. Baiocchi,et al.  Instrumental variable methods for causal inference , 2014, Statistics in medicine.

[6]  G. Imbens.  Instrumental Variables: An Econometrician's Perspective , 2014, SSRN Electronic Journal.

[7]  Stephen L. Morgan,et al.  Instrumental Variables Regression , 2014 .

[8]  Peter M. Aronow,et al.  Field Experiments and the Study of Voter Turnout , 2013 .

[9]  Dylan S. Small,et al.  Near/far matching: a study design approach to instrumental variables , 2012, Health Services and Outcomes Research Methodology.

[10]  T. VanderWeele,et al.  Power and instrument strength requirements for Mendelian randomization studies using multiple genetic variants. , 2011, International journal of epidemiology.

[11]  S. Thompson,et al.  Avoiding bias from weak instruments in Mendelian randomization studies. , 2011, International journal of epidemiology.

[12]  S. Thompson,et al.  Bias in causal estimates from Mendelian randomization studies with weak instruments , 2011, Statistics in medicine.

[13]  K. Anthony.  Anticancer drugs: Think globally, act globally , 2011, Nature Reviews Drug Discovery.

[14]  Dylan S. Small,et al.  Building a Stronger Instrument in an Observational Study of Perinatal Care for Premature Infants , 2010 .

[15]  Joshua D. Angrist,et al.  Mostly Harmless Econometrics: An Empiricist's Companion , 2008 .

[16]  George Davey Smith,et al.  Mendelian randomization: Using genes as instruments for making causal inferences in epidemiology , 2008, Statistics in medicine.

[17]  E. Miguel,et al.  Economic Shocks and Civil Conflict: An Instrumental Variables Approach , 2004, Journal of Political Economy.

[18]  S. Ebrahim,et al.  Mendelian randomization: prospects, potentials, and limitations. , 2004, International journal of epidemiology.

[19]  David W. Nickerson,et al.  Getting Out the Vote in Local Elections: Results from Six Door-to-Door Canvassing Experiments , 2003 .

[20]  S. Ebrahim,et al.  'Mendelian randomization': can genetic epidemiology contribute to understanding environmental determinants of disease? , 2003, International journal of epidemiology.

[21]  Jonathan H. Wright,et al.  A Survey of Weak Instruments and Weak Identification in Generalized Method of Moments , 2002 .

[22]  Jeffrey M. Wooldridge,et al.  Solutions Manual and Supplementary Materials for Econometric Analysis of Cross Section and Panel Data , 2003 .

[23]  Paul R. Rosenbaum,et al.  Using quantile averages in matched observational studies , 1999 .

[24]  E. Lehmann.  Elements of large-sample theory , 1998 .

[25]  Christopher R. Taber,et al.  Accounting for Dropouts in Evaluations of Social Programs , 1998, Review of Economics and Statistics.

[26]  Jean-Marie Dufour,et al.  Some Impossibility Theorems in Econometrics with Applications to Structural and Dynamic Models , 1997 .

[27]  David A. Jaeger,et al.  Problems with Instrumental Variables Estimation when the Correlation between the Instruments and the Endogenous Explanatory Variable is Weak , 1995 .

[28]  Howard S. Bloom,et al.  Accounting for No-Shows in Experimental Evaluation Designs , 1984 .

[29]  D. Rubin.  Randomization Analysis of Experimental Data: The Fisher Randomization Test Comment , 1980 .

[30]  E. Lehmann,et al.  Nonparametrics: Statistical Methods Based on Ranks , 1976 .

[31]  T. W. Anderson,et al.  Estimation of the Parameters of a Single Equation in a Complete System of Stochastic Equations , 1949 .

[32]  A. Wald.  The Fitting of Straight Lines if Both Variables are Subject to Error , 1940 .