A Distributional Perspective on Pearl's Causal Hierarchy: From Marginal to Joint and Individualized Potential Outcomes

2026/01/31 by Peng Wu, Linbo Wang
#stat.OT

paper · pdf

Abstract

Pearl's causal hierarchy is a foundational lens for formulating causal questions and is most often discussed within the framework of structural causal models. We recast the hierarchy in potential outcomes language and make its information ordering operational at the level of causal estimands. Specifically, we classify an estimand according to whether it requires marginal potential outcome distributions, their joint distribution or nested cross-world quantities, or individual-level counterfactual outcomes. We apply this criterion systematically to a broad range of estimands, including cases whose classification depends on the formulation of the scientific question. We then clarify what additional assumptions are needed when an estimand depends not only on the marginal distributions of potential outcomes, but also on their unobserved joint distribution. In particular, randomization identifies the marginals, whereas monotonicity, copula restrictions, rank preservation, and partial identification restrict or characterize the remaining uncertainty about the joint distribution. The resulting framework provides a practical map from a scientific question to an estimand, the probabilistic object it requires, and the assumptions needed for identification.

Citations

• Causal Inference: A Tale of Three Frameworks
• Marginal Causal Effect Estimation with Continuous Instrumental Variables
• Identification and Estimation of Joint Potential Outcome Distributions from a Single Study
• Cross-World Assumption and Refining Prediction Intervals for Individual Treatment Effects
• The Promises of Multiple Experiments: Identifying Joint Distribution of Potential Outcomes
• Learning Counterfactual Outcomes Under Rank Preservation
• Semiparametric principal stratification analysis beyond monotonicity
• Counterfactually Fair Reinforcement Learning via Sequential Data Preprocessing
• Identifying and bounding the probability of necessity for causes of effects with ordinal outcomes
• Learning the Effect of Persuasion via Difference-In-Differences
• Regression Adjustment for Estimating Distributional Treatment Effects in Randomized Controlled Trials
• Semiparametric Efficient Inference for the Probability of Necessary and Sufficient Causation
• Semiparametric Localized Principal Stratification Analysis with Continuous Strata
• From Probability to Counterfactuals: the Increasing Complexity of Satisfiability in Pearl's Causal Hierarchy
• Policy Learning for Balancing Short-Term and Long-Term Rewards
• Distributional Treatment Effect with Latent Rank Invariance
• Quantifying Individual Risk for Binary Outcome
• On the Comparative Analysis of Average Treatment Effects Estimation via Data Combination
• Principal Stratification with Continuous Post-Treatment Variables: Nonparametric Identification and Semiparametric Estimation
• Assessing Heterogeneity of Treatment Effects
• A First Course in Causal Inference
• The synthetic instrument: From sparse association to sparse causation
• Perspectives on harm in personalized medicine
• Deep Counterfactual Estimation with Categorical Background Variables
• Identification and estimation of treatment effects on long-term outcomes in clinical trials with external observational data
• Probabilities of Causation with Nonbinary Treatment and Effect
• Personalized Decision Making -- A Conceptual Introduction
• What's the Harm? Sharp Bounds on the Fraction Negatively Affected by Treatment
• Long-term Causal Inference Under Persistent Confounding via Data Combination
• Treatment Effect Risk: Bounds and Inference
• Sensitivity Analysis of Individual Treatment Effects: A Robust Conformal Inference Approach
• Semiparametric Estimation of Long-Term Treatment Effects
• Effects of Causes and Causes of Effects
• Causal Inference on Distribution Functions
• The Promises of Parallel Outcomes
• Causal inference methods for combining randomized trials and observational studies: a review
• Causal Inference Methods for Combining Randomized Trials and Observational Studies: A Review
• Targeting for long-term outcomes
• Conformal Inference of Counterfactuals and Individual Treatment Effects
• Fair Policy Targeting
• Principal Fairness for Human and Algorithmic Decision-Making
• On the role of surrogates in the efficient estimation of treatment effects with limited outcome data
• Probabilistic Reasoning across the Causal Hierarchy
• Confounding-Robust Policy Improvement
• Identifying the Effect of Persuasion
• Counterfactual Fairness
• Bounded, efficient and multiply robust estimation of average treatment effects using instrumental variables
• Decomposing Treatment Effect Variation
• Assessing the Treatment Effect Heterogeneity with a Latent Variable
• Principal causal effect identification and surrogate endpoint evaluation by multiple trials
• Causal analysis of ordinal treatments and binary outcomes under truncation by death
• Identifying the Distribution of Treatment Effects under Support Restrictions
• Negative Controls
• Direct and Indirect Effects
• Direct and Indirect Effects
• Probabilities of Causation: Bounds and Identification
• Efficiency of Weighted Average Derivative Estimators and Index Models
• The central role of the propensity score in observational studies for causal effects
• Advancing Counterfactual Inference through Nonlinear Quantile Regression

Related

• Distribution-consistency Structural Causal Models
• Causal Inference: A Tale of Three Frameworks
• The Causal Marginal Polytope for Bounding Treatment Effects
• Causal models on probability spaces
• Identification of Probabilities of Causation: A Complete Characterization