The Annals of Statistics

Causal inference with a graphical hierarchy of interventions

Ilya Shpitser and Eric Tchetgen Tchetgen

Full-text: Access denied (no subscription detected)

We're sorry, but we are unable to provide you with the full text of this article because we are not able to identify you as a subscriber. If you have a personal subscription to this journal, then please login. If you are already logged in, then you may need to update your profile to register your subscription. Read more about accessing full-text

Abstract

Identifying causal parameters from observational data is fraught with subtleties due to the issues of selection bias and confounding. In addition, more complex questions of interest, such as effects of treatment on the treated and mediated effects may not always be identified even in data where treatment assignment is known and under investigator control, or may be identified under one causal model but not another.

Increasingly complex effects of interest, coupled with a diversity of causal models in use resulted in a fragmented view of identification. This fragmentation makes it unnecessarily difficult to determine if a given parameter is identified (and in what model), and what assumptions must hold for this to be the case. This, in turn, complicates the development of estimation theory and sensitivity analysis procedures.

In this paper, we give a unifying view of a large class of causal effects of interest, including novel effects not previously considered, in terms of a hierarchy of interventions, and show that identification theory for this large class reduces to an identification theory of random variables under interventions from this hierarchy. Moreover, we show that one type of intervention in the hierarchy is naturally associated with queries identified under the Finest Fully Randomized Causally Interpretable Structure Tree Graph (FFRCISTG) model of Robins (via the extended g-formula), and another is naturally associated with queries identified under the Non-Parametric Structural Equation Model with Independent Errors (NPSEM-IE) of Pearl, via a more general functional we call the edge g-formula.

Our results motivate the study of estimation theory for the edge g-formula, since we show it arises both in mediation analysis, and in settings where treatment assignment has unobserved causes, such as models associated with Pearl’s front-door criterion.

Article information

Source
Ann. Statist., Volume 44, Number 6 (2016), 2433-2466.

Dates
Received: November 2014
Revised: October 2015
First available in Project Euclid: 23 November 2016

Permanent link to this document
https://projecteuclid.org/euclid.aos/1479891624

Digital Object Identifier
doi:10.1214/15-AOS1411

Mathematical Reviews number (MathSciNet)
MR3576550

Zentralblatt MATH identifier
1361.62029

Subjects
Primary: 62H05: Characterization and structure theory 62H99: None of the above, but in this section

Keywords
Causal inference graphical models mediation analysis identification

Citation

Shpitser, Ilya; Tchetgen Tchetgen, Eric. Causal inference with a graphical hierarchy of interventions. Ann. Statist. 44 (2016), no. 6, 2433--2466. doi:10.1214/15-AOS1411. https://projecteuclid.org/euclid.aos/1479891624


Export citation

References

  • [1] Avin, C., Shpitser, I. and Pearl, J. (2005). Identifiability of path-specific effects. In Proceedings of the International Joint Conference on Artificial Intelligence 19 357–363. Morgan Kaufmann, San Francisco, CA.
  • [2] Balke, A. and Pearl, J. (1994). Probabilistic evaluation of counterfactual queries. In Procedings of the Conference on Artificial Intelligence 12 230–237. Morgan Kaufmann, San Francisco, CA.
  • [3] Baron, R. M. and Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychology research: Conceptual, strategic, and statistical considerations. J. Pers. Soc. Psychol. 51 1173–1182.
  • [4] Hubbard, A. E. and van der Laan, M. J. (2008). Population intervention models in causal inference. Biometrika 95 35–47.
  • [5] Imai, K., Tingley, D. and Yamamoto, T. (2013). Experimental designs for identifying causal mechanisms. J. Roy. Statist. Soc. Ser. A 176 5–51.
  • [6] Moodie, E. E. M., Richardson, T. S. and Stephens, D. A. (2007). Demystifying optimal dynamic treatment regimes. Biometrics 63 447–455.
  • [7] Murphy, S. A. (2003). Optimal dynamic treatment regimes. J. R. Stat. Soc. Ser. B Stat. Methodol. 65 331–366.
  • [8] Neyman, J. (1990). On the application of probability theory to agricultural experiments. Essay on principles. Section 9. Statist. Sci. 5 463–472.
  • [9] Pearl, J. (1988). Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann, San Mateo, CA.
  • [10] Pearl, J. (2001). Direct and indirect effects. In Proceedings of the Conference on Uncertainty in Artificial Intelligence 17 411–420. Morgan Kaufmann, San Francisco, CA.
  • [11] Pearl, J. (2009). Causality: Models, Reasoning, and Inference, 2nd ed. Cambridge Univ. Press, Cambridge.
  • [12] Pearl, J. (2011). The causal mediation formula—A guide to the assessment of pathways and mechanisms. Technical Report R-379, Cognitive Systems Laboratory, Univ. California, Los Angeles.
  • [13] Richardson, T. S. and Robins, J. M. (2013). Single world intervention graphs (SWIGs): A unification of the counterfactual and graphical approaches to causality. Preprint. Available at http://www.csss.washington.edu/Papers/wp128.pdf.
  • [14] Robins, J. (1986). A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect. Math. Modelling 7 1393–1512.
  • [15] Robins, J. M. (1987). A graphical approach to the identification and estimation of causal parameters in mortality studies with sustained exposure periods. J. Chronic. Dis. 40 139–161.
  • [16] Robins, J. M. and Greenland, S. (1992). Identifiability and exchangeability of direct and indirect effects. Epidemiology 3 143–155.
  • [17] Robins, J. M., Hernán, M. and Brumback, B. (2000). Marginal structural models and causal inference in epidemiology. Epidemiology 11 550–560.
  • [18] Robins, J. M., Hernán, M. A. and Siebert, U. (2004). Effects of multiple interventions. In Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors 2 2191–2230. World Health Organization, Geneva.
  • [19] Robins, J. M. and Richardson, T. S. (2010). Alternative graphical causal models and the identification of direct effects. In Causality and Psychopathology: Finding the Determinants of Disorders and Their Cures (P. Shrout, ed.). Oxford University Press, Oxford.
  • [20] Robins, J. M., Rotnitzky, A. and Zhao, L. P. (1994). Estimation of regression coefficients when some regressors are not always observed. J. Amer. Statist. Assoc. 89 846–866.
  • [21] Rubin, D. B. (1974). Estimating causal effects of treatments in randomized and non-randomized studies. J. Educ. Psychol. 66 688–701.
  • [22] Rubin, D. B. (2004). Direct and indirect causal effects via potential outcomes. Scand. J. Stat. 31 161–170.
  • [23] Rubin, D. B. (2005). Causal inference using potential outcomes: Design, modeling, decisions. J. Amer. Statist. Assoc. 100 322–331.
  • [24] Shpitser, I. (2013). Counterfactual graphical models for longitudinal mediation analysis with unobserved confounding. Cognitive Science (Rumelhart Special Issue) 37 1011–1035.
  • [25] Shpitser, I. and Pearl, J. (2006). Identification of joint interventional distributions in recursive semi-Markovian causal models. In National Conference on Artificial Intelligence 21. AAAI Press, Palo Alto, CA.
  • [26] Shpitser, I. and Pearl, J. (2006). Identification of conditional interventional distributions. In Uncertainty in Artificial Intelligence 22. AUAI Press, Corvallis, OR.
  • [27] Shpitser, I. and Pearl, J. (2008). Complete identification methods for the causal hierarchy. J. Mach. Learn. Res. 9 1941–1979.
  • [28] Shpitser, I. and Pearl, J. (2009). Effects of treatment on the treated: Identification and generalization. In Uncertainty in Artificial Intelligence 25. AUAI Press, Corvallis, OR.
  • [29] Shpitser, I. and Tchetgen Tchetgen, E. (2015). Supplement to “Causal inference with a graphical hierarchy of interventions.” DOI:10.1214/15-AOS1411SUPP.
  • [30] Spirtes, P., Glymour, C. and Scheines, R. (1993). Causation, Prediction, and Search. Lecture Notes in Statistics 81. Springer, New York.
  • [31] Tchetgen Tchetgen, E. J. and Shpitser, I. (2012). Semiparametric theory for causal mediation analysis: Efficiency bounds, multiple robustness and sensitivity analysis. Ann. Statist. 40 1816–1845.
  • [32] Tchetgen Tchetgen, E. J. and Shpitser, I. (2014). Estimation of a semiparametric natural direct effect model incorporating baseline covariates. Biometrika 101 849–864.
  • [33] Tian, J. and Pearl, J. (2002). On the testable implications of causal models with hidden variables. In Uncertainty in Artificial Intelligence 18 519–527. AUAI Press, Corvallis, OR.
  • [34] Verma, T. S. and Pearl, J. (1990). Equivalence and synthesis of causal models. Technical Report R-150, Dept. Computer Science, Univ. California, Los Angeles.
  • [35] Wright, S. (1921). Correlation and causation. J. Agric. Res. 20 557–585.
  • [36] Young, J. G., Herńan, M. A. and Robins, J. M. (2014). Identification, estimation and approximation of risk under interventions that depend on the natural value of treatment using observational data. Epidemiol Method 3 1–19.

Supplemental materials

  • Supplement to “Causal inference with a graphical hierarchy of interventions”. Our supplementary materials contain detailed arguments for most of our claims, and some auxiliary definitions. In addition, we provide a detailed rationale for the use of path interventions.