Relaxed covariate overlap and margin‐based causal effect estimation

In most nonrandomized observational studies, differences between treatment groups may arise not only due to the treatment but also because of the effect of confounders. Therefore, causal inference regarding the treatment effect is not as straightforward as in a randomized trial. To adjust for confounding due to measured covariates, a variety of methods based on the potential outcomes framework are used to estimate average treatment effects. One of the key assumptions is treatment positivity, which states that the probability of treatment is bounded away from zero and one for any possible combination of the confounders. Methods for performing causal inference when this assumption is violated are relatively limited. In this article, we discuss a new balance-related condition involving the convex hulls of treatment groups, which I term relaxed covariate overlap. An advantage of this concept is that it can be linked to a concept from machine learning, termed the margin. Introduction of relaxed covariate overlap leads to an approach in which one can perform causal inference in a three-step manner. The methodology is illustrated with two examples.

[1]  Gary King,et al.  Matching as Nonparametric Preprocessing for Reducing Model Dependence in Parametric Causal Inference , 2007, Political Analysis.

[2]  B. Hansen.  Full Matching in an Observational Study of Coaching for the SAT , 2004 .

[3]  D. Ghosh,et al.  On estimating regression-based causal effects using sufficient dimension reduction , 2017 .

[4]  Debashis Ghosh,et al.  A Boosting Algorithm for Estimating Generalized Propensity Scores with Continuous Treatments , 2015, Journal of causal inference.

[5]  Dennis L. Sun,et al.  Exact post-selection inference, with application to the lasso , 2013, 1311.6238.

[6]  Mark J van der Laan,et al.  An Application of Collaborative Targeted Maximum Likelihood Estimation in Causal Inference and Genomics , 2010, The international journal of biostatistics.

[7]  J. Zubizarreta.  Stable Weights that Balance Covariates for Estimation With Incomplete Outcome Data , 2015 .

[8]  Debashis Ghosh,et al.  Penalized regression procedures for variable selection in the potential outcomes framework , 2015, Statistics in medicine.

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

[10]  D. Rubin,et al.  The central role of the propensity score in observational studies for causal effects , 1983 .

[11]  Kristin E. Porter,et al.  Diagnosing and responding to violations in the positivity assumption , 2012, Statistical methods in medical research.

[12]  Debashis Ghosh,et al.  Estimating controlled direct effects of restrictive feeding practices in the ‘Early dieting in girls’ study , 2016, Journal of the Royal Statistical Society. Series C, Applied statistics.

[13]  G. Ridgeway.  The State of Boosting ∗ , 1999 .

[14]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[15]  G. King,et al.  Multivariate Matching Methods That Are Monotonic Imbalance Bounding , 2011 .

[16]  D. Rubin.  The design versus the analysis of observational studies for causal effects: parallels with the design of randomized trials , 2007, Statistics in medicine.

[17]  J. Robins,et al.  Toward a curse of dimensionality appropriate (CODA) asymptotic theory for semi-parametric models. , 1997, Statistics in medicine.

[18]  M. J. Laan,et al.  Targeted Learning: Causal Inference for Observational and Experimental Data , 2011 .

[19]  Elizabeth A Stuart,et al.  Matching methods for causal inference: A review and a look forward. , 2010, Statistical science : a review journal of the Institute of Mathematical Statistics.

[20]  S. Kruger.  Design Of Observational Studies , 2016 .

[21]  Michael Rosenblum,et al.  Marginal Structural Models , 2011 .

[22]  Dylan S. Small,et al.  Defining the Study Population for an Observational Study to Ensure Sufficient Overlap: A Tree Approach , 2011 .

[23]  R. Schneider.  Convex Bodies: The Brunn–Minkowski Theory: Minkowski addition , 1993 .

[24]  K. Imai,et al.  Covariate balancing propensity score , 2014 .

[25]  Alexander D'Amour,et al.  Overlap in observational studies with high-dimensional covariates , 2017, Journal of Econometrics.

[26]  Nello Cristianini,et al.  An Introduction to Support Vector Machines and Other Kernel-based Learning Methods , 2000 .

[27]  K. Schittkowski,et al.  NONLINEAR PROGRAMMING , 2022 .

[28]  K. C. G. Chan,et al.  Globally efficient non‐parametric inference of average treatment effects by empirical balancing calibration weighting , 2016, Journal of the Royal Statistical Society. Series B, Statistical methodology.

[29]  G. Wahba.  Spline models for observational data , 1990 .

[30]  Marc T. Ratkovic.  Balancing within the Margin: Causal Effect Estimation with Support Vector Machines∗ , 2014 .

[31]  Bernard Chazelle,et al.  An optimal convex hull algorithm in any fixed dimension , 1993, Discret. Comput. Geom..

[32]  Samy Suissa,et al.  The positivity assumption and marginal structural models: the example of warfarin use and risk of bleeding , 2012, European Journal of Epidemiology.

[33]  J. Lunceford,et al.  Stratification and weighting via the propensity score in estimation of causal treatment effects: a comparative study , 2004, Statistics in medicine.

[34]  D B Rubin,et al.  Practical implications of modes of statistical inference for causal effects and the critical role of the assignment mechanism. , 1991, Biometrics.

[35]  Richard K. Crump,et al.  Dealing with limited overlap in estimation of average treatment effects , 2009 .

[36]  D. Rubin.  Estimating causal effects of treatments in randomized and nonrandomized studies. , 1974 .

[37]  L. Birch,et al.  Eating in the absence of hunger and overweight in girls from 5 to 7 y of age. , 2002, The American journal of clinical nutrition.

[38]  J. Sekhon,et al.  Genetic Matching for Estimating Causal Effects: A General Multivariate Matching Method for Achieving Balance in Observational Studies , 2006, Review of Economics and Statistics.

[39]  Nandita Mitra,et al.  Radiation therapy is associated with improved survival in the adjuvant and definitive treatment of intrahepatic cholangiocarcinoma. , 2008, International journal of radiation oncology, biology, physics.

[40]  Jinbo Bi,et al.  A geometric approach to support vector regression , 2003, Neurocomputing.

[41]  Kosuke Imai,et al.  Causal Inference With General Treatment Regimes , 2004 .

[42]  G. Imbens.  The Role of the Propensity Score in Estimating Dose-Response Functions , 1999 .