Foundations of Science

, Volume 3, Issue 1, pp 151–182 | Cite as

An Axiomatic Characterization of Causal Counterfactuals

  • David Galles
  • Judea Pearl

Abstract

This paper studies the causal interpretation of counterfactual sentences using a modifiable structural equation model. It is shown that two properties of counterfactuals, namely, composition and effectiveness, are sound and complete relative to this interpretation, when recursive (i.e., feedback-less) models are considered. Composition and effectiveness also hold in Lewis's closest-world semantics, which implies that for recursive models the causal interpretation imposes no restrictions beyond those embodied in Lewis's framework. A third property, called reversibility, holds in nonrecursive causal models but not in Lewis's closest-world semantics, which implies that Lewis's axioms do not capture some properties of systems with feedback. Causal inferences based on counterfactual analysis are exemplified and compared to those based on graphical models.

causality counterfactuals interventions structural equations policy analysis graphical models 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. J.D. Angrist, G.W. Imbens, and Rubin D.B. (1996), ‘Identification of causal effects using instrumental variables (with comments).’ Journal of the American Statistical Association, 91(434):444–472, June.CrossRefGoogle Scholar
  2. A. Balke and J. Pearl (1994), ‘Counterfactual probabilities: Computation methods, bounds, and applications’, in: R.L. de Mantaras and D. Poole (eds.), Proceedings of the Tenth Conference on Uncertainty in Artificial Intelligence, pages 11–18, San Francisco, Morgan Kaufmann.Google Scholar
  3. A. Balke and J. Pearl (1995), ‘Counterfactuals and policy analysis in structural models’ in: P. Besnard and S. Hanks, (eds.), Proceedings of the Eleventh Conference on Uncertainty in Artificial Intelligence, pages 11–18, San Francisco, Morgan Kaufmann.Google Scholar
  4. R.J. Bowden and D.A. Turkington (1984), Instrumental Variables. Cambridge University Press, Cambridge, MA.Google Scholar
  5. A.P. Dawid (1997), Causal inference without counterfactuals. Technical Report, Department of Statistical Science, University College London, UK.Google Scholar
  6. P.J. Dhrymcs (1970), Economctrics. Springer-Verlag, New York.Google Scholar
  7. R.F. Engle, D.F. Hendry, and J.F. Richard (1983), ‘Exogeneity’. Econometrical, 51(2):277–304, March.CrossRefGoogle Scholar
  8. R. Fagin, J.M. Halpert, Y. Moses, and M.Y. Vardi (1995), Reasoning About Knowledge. MIT Press, Cambridge, MA.Google Scholar
  9. F.M. Fisher (1966), The Identification Problem in Econometrics. McGraw-Hill, New York.Google Scholar
  10. F.M. Fisher (1970), ‘A correspondence principle for simultaneous equation models’. Econometrica, 38:73–92.CrossRefGoogle Scholar
  11. D. Galles and J. Pearl (1997a) An axiomatic characterization of causal counterfactuals. Technical Report R-250-L, Computer Science Department, University of California, Los Angeles, March 1997. Prepared for Foundations of Science, Kluwer Academic Publishers.Google Scholar
  12. D. Galles and J. Pearl (1997b), ‘Axioms of causal relevance’, Artificial Intelligence, 97(1–2):9–43.CrossRefGoogle Scholar
  13. A. Gibbard and L. Harper (1981), ‘Counterfactuals and two kinds of expected utility’, in: W.L. Harper, R. Stalnaker, and G. Pearce, (eds.), Ifs. D. Reidel, Dordrecht: Holland.Google Scholar
  14. M.L. Ginsberg and D.E. Smith (1987), ‘Reasoning about action I: A possible worlds approach’, In Frank M. Brown, (ed.), The Frame Problem in Artificial Intelligence, pages 233–258. Morgan Kaufmann, Los Altos, CA.Google Scholar
  15. Arthur S. Goldberger (1992), ‘Models of substance [comment on N. Wermuth, “On block-recursive linear regression equations”]’. Brazilian Journal of Probability and Statistics, 6:1–56.Google Scholar
  16. J. Halpern (1997), Axiomatizing causal structures. Unpublished report, Cornell University, May.Google Scholar
  17. D. Heckerman and R. Shachter (1995), ‘A definition and graphical representation of causality’, in: P. Besnard and S. Hanks, (eds.), Proceedings of the Eleventh Conference on Uncertainty in Artificial Intelligence, pages 262–273, San Francisco, Morgan Kaufmann.Google Scholar
  18. David F. Hendry (1995), Dynamic Econometrics. Oxford University Press, New York.Google Scholar
  19. P. W. Holland (1986), ‘Statistics and causal inference (with discussion)’. Journal of the American Statistical Association, 81(396):945–970.CrossRefGoogle Scholar
  20. P.W. Holland (1988), ‘Causal inference, path analysis, and recursive structural equations models.’ In C. Clogg, editor, Sociological Methodology, pages 449–484. American Sociological Association, Washington, D.C.Google Scholar
  21. H. Katsuno and A.O. Mendelzon (1991), ‘On the difference between updating a knowledge base and revising it.’ In Principles of Knowledge Representation and Reasoning: Proceedings of the Second International Conference, pages 387–394. Boston, MA.Google Scholar
  22. T.C. Koopmans (1950), ‘When is an equation system complete for statistical purposes?’ In T.C. Koopmans, (ed.), Statistical Inference in Dynamic Economic Models, Cowles Commission, Monograph 10. Wiley, New York. Reprinted in D.F. Hendry and M.S. Morgan (Eds.), The Foundations of Econometric Analysis, Cambridge University Press, 527–537, 1995.Google Scholar
  23. E. Leamer (1985), ‘Vector autoregression for causal inference?’ Carnegie-Rochester Conference Series on Public Policy, 22:255–304.CrossRefGoogle Scholar
  24. D. Lewis (1973a), ‘Causation.’ Journal of Philosophy, 70:556–567.CrossRefGoogle Scholar
  25. D. Lewis (1973b), Counterfactuals. Harvard University Press, Cambridge, MA.Google Scholar
  26. D. Lewis (1981), ‘Counterfactuals and comparative possibility’, in: W.L. Harper, R. Stalnaker, and G. Pearce, (eds.), Ifs. D. Reidel, Dordrecht, Holland.Google Scholar
  27. C.F. Manski (1990), ‘Nonparametric bounds on treatment effects.’ American Economic Review, Papers and Proceedings, 80:319–323.Google Scholar
  28. J.S. Meditch (1969), Stochastic Optimal Linear Estimation and Control. McGraw-Hill, New York.Google Scholar
  29. J. Pearl (1988), Probabilistic Reasoning in Intelligent Systems. Morgan Kaufmann San Mateo, CA. (Revised 2nd printing, 1992).Google Scholar
  30. J. Pearl (1994), ‘A probabilistic calculus of actions’, in: R.L. de Mantaras and D. Poole, (eds.), Proceedings of the Tenth Conference on Uncertainty in Artificial Intelligence, pages 454–462, San Francisco, Morgan Kaufmann.Google Scholar
  31. J. Pearl (1995a), ‘Causal diagrams for empirical research (with discussion). Biometrika, 82(4):669–709.CrossRefGoogle Scholar
  32. J. Pearl (1995b), On the testability of causal models with latent and instrumental variables.’ In P. Besnard and S. Hanks, (eds), Uncertainty in Artificial Intelligence 11, pages 435–443. Morgan Kaufmann.Google Scholar
  33. J. Pearl (1996), ‘Structural and probabilistic causality.’ Psychology of Learning and Motivation, 34:393–435.CrossRefGoogle Scholar
  34. J. Pearl (1997a), “The new challenge: From a century of statistics to an age of causation.’ Computing Science and Statistics, 29(2):415–423.Google Scholar
  35. J. Pearl (1997b), ‘On the identification of nonparametric structural models.’ In M. Berkane, (ed.), Latent Variable Modeling with Application to Causality, pages 29–68. Springer-Verlag.Google Scholar
  36. J. Pearl (1998a), Graphs, causality, and structural equation models. Technical Report R-253, Department of Computer Science, University of California, Los Angeles. To appear in Sociological Methods and Research, Special Issue on Causality.Google Scholar
  37. J. Pearl (1998b), Why there is no statistical test for confounding, why many think there is, and why they are almost right. Technical Report R-256, Department of Computer Science, University of California, Los Angeles.Google Scholar
  38. J.W. Pratt and R. Schlaifer (1988), ‘On the interpretation and observation of laws.’ Journal of Econometrics, 39:23–52.CrossRefGoogle Scholar
  39. J.F. Richard (1980), ‘Models with several regimes and charges in exogeneity.’ The Review of Economic Studies, 47:1–20.CrossRefGoogle Scholar
  40. J. Robins (1986), ‘A new approach to causal inference in mortality studies with a sustained exposure period — applications to control of the healthy workers survivor effect.’ Mathematical Modeling, 7:1393–512.CrossRefGoogle Scholar
  41. J. Robins (1987), ‘Addendum to “A new approach to causal inference in mortality studies with sustained exposure periods — application to control of the healthy worker survivor effect”.’ Computers and Mathematics, with Applications., 14:923–45.CrossRefGoogle Scholar
  42. J.M. Robins (1995), ‘Discussion of “Causal diagrams for empirical research” by J. Pearl.’ Biometrika, 82(4):695–698.CrossRefGoogle Scholar
  43. P. Rosenbaum and D. Rubin (1983), ‘The central role of propensity score in observational studies for causal effects.’ Biometrika, 70:41–55.CrossRefGoogle Scholar
  44. D.B. Rubin (1974), ‘Estimating causal effects of treatments in randomized and nonrandomized studies.’ Journal of Educational Psychology, 66:688–701.CrossRefGoogle Scholar
  45. H.A. Simon and N. Rescher (1966), ‘Cause and counterfactual.’ Philosophy and Science, 33:323–340.CrossRefGoogle Scholar
  46. M.E. Sobel (1990), ‘Effect analysis and causation in linear structural equation models.’ Psychometrika, 55:495–515.CrossRefGoogle Scholar
  47. P. Spirtes, C. Glymour, and R. Schienes (1993), Causation, Prediction, and Search. Springer-Verlag, New York.Google Scholar
  48. R.H. Strotz and O.A. Wold (1960), ‘Recursive versus nonrecursive systems: At attempt at synthesis.’ Econometrica, 28:417–427.CrossRefGoogle Scholar
  49. M. Winslett (1988), ‘Reasoning about action using a possible worlds approach. In Proceedings of the Seventh American Association for Artificial Intelligenc Conference, pages 89–93.Google Scholar

Copyright information

© Kluwer Academic Publishers 1998

Authors and Affiliations

  • David Galles
    • 1
  • Judea Pearl
    • 1
  1. 1.Cognitive Systems Laboratory Computer Science DepartmentUniversity of CaliforniaLos Angeles

Personalised recommendations