External validity: from do-calculus to transportability across populations
From MaRDI portal
Publication:252808
DOI10.1214/14-STS486zbMATH Open1331.62326arXiv1503.01603WikidataQ57258911 ScholiaQ57258911MaRDI QIDQ252808FDOQ252808
Authors: Judea Pearl, Elias Bareinboim 
Publication date: 4 March 2016
Published in: Statistical Science (Search for Journal in Brave)
Abstract: The generalizability of empirical findings to new environments, settings or populations, often called "external validity," is essential in most scientific explorations. This paper treats a particular problem of generalizability, called "transportability," defined as a license to transfer causal effects learned in experimental studies to a new population, in which only observational studies can be conducted. We introduce a formal representation called "selection diagrams" for expressing knowledge about differences and commonalities between populations of interest and, using this representation, we reduce questions of transportability to symbolic derivations in the do-calculus. This reduction yields graph-based procedures for deciding, prior to observing any data, whether causal effects in the target population can be inferred from experimental findings in the study population. When the answer is affirmative, the procedures identify what experimental and observational findings need be obtained from the two populations, and how they can be combined to ensure bias-free transport.
Full work available at URL: https://arxiv.org/abs/1503.01603
Recommendations
- External Validity: From Do -Calculus to Transportability Across Populations
- A simple approximation for evaluating external validity bias
- Validity in a dialetheist framework
- Influence of calibration and validation sets' similarity on the result of external validation test
- Expressing validity: towards a self-sufficient inferentialism
- Derivability and metainferential validity
General nonlinear regression (62J02) Design of statistical experiments (62Kxx) Multivariate analysis (62Hxx)
Cites Work
- Causation, prediction, and search
- Title not available (Why is that?)
- Title not available (Why is that?)
- Sample Selection Bias as a Specification Error
- Title not available (Why is that?)
- A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect
- Causality. Models, reasoning, and inference
- Influence Diagrams for Causal Modelling and Inference
- Title not available (Why is that?)
- Karl Pearson's meta-analysis revisited
- Settable systems: an extension of Pearl's causal model with optimization, equilibrium, and learning
- The structural theory of causation
- Related Causal Frameworks for Surrogate Outcomes
- Causal diagrams for empirical research
- Markov Properties for Acyclic Directed Mixed Graphs
- TRYGVE HAAVELMO AND THE EMERGENCE OF CAUSAL CALCULUS
- The Statistical Implications of a System of Simultaneous Equations
- Causal inference in statistics: an overview
Cited In (36)
- Graphical models for processing missing data
- Model transfer across additive manufacturing processes via mean effect equivalence of lurking variables
- Title not available (Why is that?)
- Surrogate outcomes and transportability
- Developments in Survey Research over the Past 60 Years: A Personal Perspective
- Assumption-Lean Cox Regression
- Bayesian data fusion: Probabilistic sensitivity analysis for unmeasured confounding using informative priors based on secondary data
- Regression analysis of unmeasured confounding
- Transportability of causal inference under random dynamic treatment regimes for kidney-pancreas transplantation
- Lurking variable detection via dimensional analysis
- Efficient, doubly robust estimation of the effect of dose switching for switchers in a randomized clinical trial
- Transporting stochastic direct and indirect effects to new populations
- Discussion of Kallus and Mo, Qi, and Liu: New Objectives for Policy Learning
- Rejoinder: Learning Optimal Distributionally Robust Individualized Treatment Rules
- Data integration in causal inference
- Efficient and Robust Methods for Causally Interpretable Meta-Analysis: Transporting Inferences from Multiple Randomized Trials to a Target Population
- Leveraging population outcomes to improve the generalization of experimental results: application to the JTPA study
- Learning optimal distributionally robust individualized treatment rules
- From Local to Global: External Validity in a Fertility Natural Experiment
- Surrogacy validation for time-to-event outcomes with illness-death frailty models
- Generalizing the Results from Social Experiments: Theory and Evidence from India
- Fusion designs and estimators for treatment effects
- From statistical to causal learning
- A calibration approach to transportability and data-fusion with observational data
- Comment on: ``Target estimands for population-adjusted indirect comparisons
- Target estimands for population-adjusted indirect comparisons
- Efficient nonparametric inference on the effects of stochastic interventions under two‐phase sampling, with applications to vaccine efficacy trials
- Causal structure learning: a combinatorial perspective
- An empirical study of testing independencies in Bayesian networks using rp-separation
- 2017 Klein lecture: the science of using science: toward an understanding of the threats to scalability
- Causal inference methods for combining randomized trials and observational studies: a review
- Extending inferences from a randomized trial to a new target population
- Transporting experimental results with entropy balancing
- Inferring network structure from interventional time-course experiments
- Complete graphical characterization and construction of adjustment sets in Markov equivalence classes of ancestral graphs
- Doubly robust estimation of the weighted average treatment effect for a target population
Uses Software
This page was built for publication: External validity: from do-calculus to transportability across populations
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q252808)
