Semiparametric Inference For Causal Effects In Graphical Models With Hidden Variables

The last decade witnessed the development of algorithms that completely solve the identifiability problem for causal effects in hidden variable causal models associated with directed acyclic graphs. However, much of this machinery remains underutilized in practice owing to the complexity of estimating identifying functionals yielded by these algorithms. In this paper, we provide simple graphical criteria and semiparametric estimators that bridge the gap between identification and estimation for causal effects involving a single treatment and a single outcome. First, we provide influence function based doubly robust estimators that cover a significant subset of hidden variable causal models where the effect is identifiable. We further characterize an important subset of this class for which we demonstrate how to derive the estimator with the lowest asymptotic variance, i.e., one that achieves the semiparametric efficiency bound. Finally, we provide semiparametric estimators for any single treatment causal effect parameter identified via the aforementioned algorithms. The resulting estimators resemble influence function based estimators that are sequentially reweighted, and exhibit a partial double robustness property, provided the parts of the likelihood corresponding to a set of weight models are correctly specified. Our methods are easy to implement and we demonstrate their utility through simulations.

[1]  P. Spirtes,et al.  MARKOV EQUIVALENCE FOR ANCESTRAL GRAPHS , 2009, 0908.3605.

[2]  James M. Robins,et al.  On the Validity of Covariate Adjustment for Estimating Causal Effects , 2010, UAI.

[3]  Tom Burr,et al.  Causation, Prediction, and Search , 2003, Technometrics.

[4]  W. Newey,et al.  Semiparametric Efficiency Bounds , 1990 .

[5]  Robert E. Tarjan,et al.  Depth-First Search and Linear Graph Algorithms , 1972, SIAM J. Comput..

[6]  Judea Pearl,et al.  Identification of Joint Interventional Distributions in Recursive Semi-Markovian Causal Models , 2006, AAAI.

[7]  Eric J. Tchetgen Tchetgen,et al.  Robust inference on indirect causal effects , 2017 .

[8]  M. Maathuis,et al.  Graphical criteria for efficient total effect estimation via adjustment in causal linear models , 2019, 1907.02435.

[9]  L. Gagliardi,et al.  Directed Acyclic Graphs: a Tool for Causal Studies in Pediatrics , 2018, Pediatric Research.

[10]  J. Pearl,et al.  Causal diagrams for epidemiologic research. , 1999, Epidemiology.

[11]  T. Richardson.  Single World Intervention Graphs ( SWIGs ) : A Unification of the Counterfactual and Graphical Approaches to Causality , 2013 .

[12]  R. Evans.  Margins of discrete Bayesian networks , 2015, The Annals of Statistics.

[13]  P. Spirtes,et al.  Causation, prediction, and search , 1993 .

[14]  James M. Robins,et al.  Nested Markov Properties for Acyclic Directed Mixed Graphs , 2012, UAI.

[15]  J. Robins.  A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect , 1986 .

[16]  Ezequiel Smucler,et al.  Efficient adjustment sets for population average treatment effect estimation in non-parametric causal graphical models , 2019, ArXiv.

[17]  Tianchen Qian,et al.  Deductive derivation and turing‐computerization of semiparametric efficient estimation , 2015, Biometrics.

[18]  J. Robins,et al.  Doubly Robust Estimation in Missing Data and Causal Inference Models , 2005, Biometrics.

[19]  J. Hájek.  A characterization of limiting distributions of regular estimates , 1970 .

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

[21]  J. Robins,et al.  Estimation of Regression Coefficients When Some Regressors are not Always Observed , 1994 .

[22]  Marco Valtorta,et al.  Pearl's Calculus of Intervention Is Complete , 2006, UAI.

[23]  Elias Bareinboim,et al.  Estimating Causal Effects Using Weighting-Based Estimators , 2020, AAAI.

[24]  James M. Robins,et al.  Marginal Structural Models versus Structural nested Models as Tools for Causal inference , 2000 .

[25]  Eric J. Tchetgen Tchetgen,et al.  Robust inference on population indirect causal effects: the generalized front door criterion , 2017, Journal of the Royal Statistical Society. Series B, Statistical methodology.

[26]  George B. Dantzig,et al.  A PRIMAL--DUAL ALGORITHM , 1956 .

[27]  K. Do,et al.  Efficient and Adaptive Estimation for Semiparametric Models. , 1994 .

[28]  T. Richardson.  Markov Properties for Acyclic Directed Mixed Graphs , 2003 .

[29]  James M. Robins,et al.  INTRODUCTION TO NESTED MARKOV MODELS , 2014 .

[30]  Maciej Liskiewicz,et al.  Efficiently Finding Conditional Instruments for Causal Inference , 2015, IJCAI.

[31]  Jin Tian,et al.  A general identification condition for causal effects , 2002, AAAI/IAAI.

[32]  M. Drton.  Discrete chain graph models , 2009, 0909.0843.

[33]  P. Bickel.  Efficient and Adaptive Estimation for Semiparametric Models , 1993 .

[34]  P. Spirtes,et al.  Ancestral graph Markov models , 2002 .

[35]  Michael I. Jordan.  Graphical Models , 2003 .

[36]  Marco Carone,et al.  Toward Computerized Efficient Estimation in Infinite-Dimensional Models , 2016, Journal of the American Statistical Association.

[37]  D. Luenberger.  Optimization by Vector Space Methods , 1968 .

[38]  Elias Bareinboim,et al.  Causal Inference and Data Fusion in Econometrics , 2019, The Econometrics Journal.

[39]  Thomas S. Richardson,et al.  Acyclic Linear SEMs Obey the Nested Markov Property , 2018, UAI.

[40]  Edward H Kennedy,et al.  Non‐parametric methods for doubly robust estimation of continuous treatment effects , 2015, Journal of the Royal Statistical Society. Series B, Statistical methodology.

[41]  J. Hahn.  On the Role of the Propensity Score in Efficient Semiparametric Estimation of Average Treatment Effects , 1998 .

[42]  Thomas Richardson,et al.  A Potential Outcomes Calculus for Identifying Conditional Path-Specific Effects , 2019, AISTATS.

[43]  J. Pearl.  Causal diagrams for empirical research , 1995 .

[44]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems , 1988 .

[45]  Motoaki Kawanabe,et al.  Dimensionality reduction for density ratio estimation in high-dimensional spaces , 2010, Neural Networks.

[46]  A. Tsiatis.  Semiparametric Theory and Missing Data , 2006 .

[47]  Jin Tian,et al.  On the Testable Implications of Causal Models with Hidden Variables , 2002, UAI.

[48]  F. Hampel.  The Influence Curve and Its Role in Robust Estimation , 1974 .

[49]  Judea Pearl,et al.  Equivalence and Synthesis of Causal Models , 1990, UAI.

[50]  Johannes Textor,et al.  A Complete Generalized Adjustment Criterion , 2015, UAI.

[51]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[52]  Thomas S. Richardson,et al.  Maximum likelihood fitting of acyclic directed mixed graphs to binary data , 2010, UAI.

[53]  Thomas S. Richardson,et al.  Smooth, identifiable supermodels of discrete DAG models with latent variables , 2015, Bernoulli.

[54]  Zhao Hui Zheng Zhongguo Liu Baijun,et al.  On the Markov equivalence of maximal ancestral graphs , 2005 .

[55]  J. Robins,et al.  Double/Debiased Machine Learning for Treatment and Structural Parameters , 2017 .