Adaptive Hyper-box Matching for Interpretable Individualized Treatment Effect Estimation

We propose a matching method for observational data that matches units with others in unit-specific, hyper-box-shaped regions of the covariate space. These regions are large enough that many matches are created for each unit and small enough that the treatment effect is roughly constant throughout. The regions are found as either the solution to a mixed integer program, or using a (fast) approximation algorithm. The result is an interpretable and tailored estimate of a causal effect for each unit.

[1]  D. Freedman,et al.  On the histogram as a density estimator:L2 theory , 1981 .

[2]  Cynthia Rudin,et al.  MALTS: Matching After Learning to Stretch , 2018, J. Mach. Learn. Res..

[3]  D. Rubin.  Causal Inference Using Potential Outcomes , 2005 .

[4]  M. Wand.  Data-Based Choice of Histogram Bin Width , 1997 .

[5]  G. King,et al.  Causal Inference without Balance Checking: Coarsened Exact Matching , 2012, Political Analysis.

[6]  Illtyd Trethowan.  Causality , 1938 .

[7]  Nicholas I. Fisher,et al.  Bump hunting in high-dimensional data , 1999, Stat. Comput..

[8]  P. Richard Hahn,et al.  Bayesian Regression Tree Models for Causal Inference: Regularization, Confounding, and Heterogeneous Effects , 2017, 1706.09523.

[9]  Lazaros G. Papageorgiou,et al.  A mixed integer optimisation model for data classification , 2009, Comput. Ind. Eng..

[10]  J. Zubizarreta,et al.  Evaluation of subset matching methods and forms of covariate balance , 2016, Statistics in medicine.

[11]  Rajeev Dehejia,et al.  Propensity Score-Matching Methods for Nonexperimental Causal Studies , 2002, Review of Economics and Statistics.

[12]  Christian Hansen,et al.  Double/Debiased/Neyman Machine Learning of Treatment Effects , 2017, 1701.08687.

[13]  Paul R. Rosenbaum,et al.  Matching for Balance, Pairing for Heterogeneity in an Observational Study of the Effectiveness of For-Profit and Not-For-Profit High Schools in Chile , 2014, 1404.3584.

[14]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[15]  H. Chipman,et al.  BART: Bayesian Additive Regression Trees , 2008, 0806.3286.

[16]  Samuel D. Pimentel,et al.  Optimal multilevel matching using network flows: An application to a summer reading intervention , 2018 .

[17]  Jared S. Murray,et al.  Bayesian Additive Regression Trees: A Review and Look Forward , 2020, Annual Review of Statistics and Its Application.

[18]  Paul R. Rosenbaum,et al.  Imposing Minimax and Quantile Constraints on Optimal Matching in Observational Studies , 2017 .

[19]  L. Schmetterer.  Zeitschrift fur Wahrscheinlichkeitstheorie und Verwandte Gebiete. , 1963 .

[20]  Jennifer L. Hill,et al.  Bayesian Nonparametric Modeling for Causal Inference , 2011 .

[21]  J. Zubizarreta.  Journal of the American Statistical Association Using Mixed Integer Programming for Matching in an Observational Study of Kidney Failure after Surgery Using Mixed Integer Programming for Matching in an Observational Study of Kidney Failure after Surgery , 2022 .

[22]  B. Hansen,et al.  Optimal Full Matching and Related Designs via Network Flows , 2006 .

[23]  Paul R. Rosenbaum,et al.  Optimal Matching for Observational Studies , 1989 .

[24]  Cynthia Rudin,et al.  FLAME: A Fast Large-scale Almost Matching Exactly Approach to Causal Inference , 2017, J. Mach. Learn. Res..

[25]  Georg Peters,et al.  Granular Box Regression , 2011, IEEE Transactions on Fuzzy Systems.

[26]  Md. Noor-E-Alam,et al.  Hypothesis Tests That Are Robust to Choice of Matching Method , 2018 .

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

[28]  B. Hansen.  The prognostic analogue of the propensity score , 2008 .

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

[30]  Cynthia Rudin,et al.  Interpretable Almost-Exact Matching for Causal Inference , 2019, AISTATS.

[31]  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.

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

[33]  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.

[34]  Stefan Wager,et al.  Estimation and Inference of Heterogeneous Treatment Effects using Random Forests , 2015, Journal of the American Statistical Association.

[35]  Tong Wang,et al.  Causal Rule Sets for Identifying Subgroups with Enhanced Treatment Effect , 2017, INFORMS J. Comput..

[36]  D. Rubin,et al.  Statistical Analysis with Missing Data. , 1989 .

[37]  Bogdan Gabrys,et al.  Hyperbox-based machine learning algorithms: a comprehensive survey , 2019, Soft Computing.

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

[39]  D. W. Scott.  On optimal and data based histograms , 1979 .

[40]  Judea Pearl,et al.  Causal Inference , 2010 .

[41]  Jared S. Murray,et al.  Model Interpretation Through Lower-Dimensional Posterior Summarization , 2019, J. Comput. Graph. Stat..

[42]  J. Pearl,et al.  Causal Inference , 2011, Twenty-one Mental Models That Can Change Policing.

[43]  Cynthia Rudin,et al.  Box drawings for learning with imbalanced data , 2014, KDD.

[44]  Elizabeth A Stuart,et al.  Prognostic score-based balance measures can be a useful diagnostic for propensity score methods in comparative effectiveness research. , 2013, Journal of clinical epidemiology.