Causal inference in statistics: An overview
Open accessThe DOI matches its Crossref record; 136 of the 136 references it cites were found in a registry or an index; 2 of the 72 sources its claims cite were checked for support.
Abstract
This review presents empirical researchers with recent advances in causal inference, and stresses the paradigmatic shifts that must be undertaken in moving from traditional statistical analysis to causal analysis of multivariate data. Special emphasis is placed on the assumptions that underly all causal inferences, the languages used in formulating those assumptions, the conditional nature of all causal and counterfactual claims, and the methods that have been developed for the assessment of such claims. These advances are illustrated using a general theory of causation based on the Structural Causal Model (SCM) described in Pearl (2000a), which subsumes and unifies other approaches to causation, and provides a coherent mathematical foundation for the analysis of causes and counterfactuals. In particular, the paper surveys the development of mathematical tools for inferring (from a combination of data and assumptions) answers to three types of causal queries: (1) queries about the effects of potential interventions, (also called “causal effects” or “policy evaluation”) (2) queries about probabilities of counterfactuals, (including assessment of “regret,” “attribution” or “causes of effects”) and (3) queries about direct and indirect effects (also known as “mediation”). Finally, the paper defines the formal and conceptual relationships between the structural and potential-outcome frameworks and presents tools for a symbiotic analysis that uses the strong features of both.
Overview
CoChat Summary
TL;DR
This survey argues that causal inference requires more than standard statistical association: it needs a formal language for causal assumptions and a mathematical machinery for combining those assumptions with data. Pearl presents the Structural Causal Model as a unifying framework that subsumes graphical, potential-outcome, and structural-equation approaches.
Key findings
- The paper distinguishes associational concepts (e.g., correlation, regression, conditional independence) from causal concepts (e.g., intervention, effect, confounding, attribution), arguing that causal claims cannot be defined from observed distributions alone.
- In Markovian structural models, the post-intervention distribution is identified by truncated factorization; for example, the paper shows that \(E(Y \mid do(x_0)) = E(Y \mid X=x_0)\) under the stated assumptions.
- The back-door criterion provides a graphical rule for choosing admissible adjustment sets; the paper states that if a set \(S\) blocks all back-door paths from \(X\) to \(Y\), then stratifying on \(S\) removes confounding bias.
- For noncompliance in clinical trials, the paper says treatment effects are generally nonidentifiable, but bounds can be derived; Balke and Pearl (1997) obtained closed-form bounds for 16 equivalent classes in the binary case, and the paper notes that these bounds can yield “significant and sometimes accurate” information.
Methods
- This is a conceptual review/survey, not an empirical study; it synthesizes advances in causal analysis “from the unifying perspective” of the Structural Causal Model.
- It develops the formalism using structural equations, causal diagrams, do-operator interventions, and counterfactual notation, and then maps these ideas to the potential-outcome framework.
- The analysis is mainly theoretical and illustrative, using examples such as mediation, instrumental variables, covariate adjustment/back-door identification, and noncompliance in clinical trials.
Limitations
- The paper states that counterfactual probabilities conditioned on outcomes “can in general [not] be identified from observational or even experimental studies” without additional conditions.
- It states that natural direct effects “cannot in general be identified or estimated, even with the help of ideal, controlled experiments” unless extra assumptions hold.
- Evident from the text: this is a survey and framework paper, so it does not report a new dataset, sample size, or empirical effect estimate.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
2 Valid · 1 Not valid · 53 Not checked · 133 Own results and interpretations
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“Note also that the truncated factorization formula is not restricted to interventions on a single variable; it is applicable to simultaneous or sequential interventions such as those invoked in the analysis of time varying treatment with time varying confounders (Robins, 1986; Arjas and Parner, 2004).”
What the sources say
- Supportedcites Robins, 1986 source
Section F describes treatments assigned at successive times, with assignment depending on past measured covariates. The G-Computation Algorithm multiplies conditional probabilities along treatment and covariate-history paths and sums across those paths to calculate outcomes under a hypothetical treatment. This supports the cited point that the calculation extends beyond an intervention on one variable, although the paper calls it the G-computation algorithm rather than the truncated factorization formula. (LLM, conf=0.87)
- Not checked — no full text of the sourcecites Arjas and Parner, 2004 source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“The selection criterion can be applied systematically to diagrams of any size and shape, thus freeing analysts from judging whether ' X is conditionally ignorable given S ,' a formidable mental task required in the potential-response framework (Rosenbaum and Rubin, 1983).”
What the sources say
- Supportedcites Rosenbaum and Rubin, 1983 source
The citation supports the sentence’s reference to conditional ignorability in the potential-response framework. Section 1-2 defines strongly ignorable treatment assignment using the independence of treatment assignment from potential responses given covariates. The diagrammatic selection criterion is the citing authors’ own contrast, not a method attributed to Rosenbaum and Rubin. (LLM, conf=0.88)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“A few but challenging algebraic steps would reveal that one can perform the summation over Z 2 to obtain”
- Also stated in: § 3.3. Coping with unmeasured confounders · plain
“which means that we need only adjust for Z 1 and Z 3 without ever measuring Z 2 .”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § 3.3. Coping with unmeasured confounders · plain
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The set { Z 3 } , however, is not sufficient for adjustment because, as explained above, it does not block the path X ← W 1 ← Z 1 → Z 3 ← Z 2 → W 2 → Y .”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“This expression coincides with Robins' (1987) G -computation formula, which was derived from a more complicated set of (counterfactual) assumptions.”
What the sources say
- Not checked — no full text of the sourcecites Robins, 1987 source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
114 more claims
Create a free account to see all 189 claims and to verify the 7 references still open.
- Probably not a substantial claim. 39
- It may state two separate claims as one. 39
- The quote only partly states this claim. 1
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“which is known as the truncated factorization formula , see for example, Corollary 1 in Pearl (2009).”
From Discrimination-Free Insurance Pricing (2022) · cites this paper as Pearl (2009)
In the citing paper
““observational data” (such as rfMRI) is in general not a robust and safe tool for inferring causality, compared with “interventional” studies (such as task-fMRI experiments) [63].”
From Functional connectomics from resting-state fMRI (2013) · cites this paper as [63]
Citations
Cites
The central role of the propensity score in observational studies for causal effects
ROSENBAUM, P. R., & RUBIN, D. B. (1983) · Biometrika · doi:10.1093/biomet/70.1.41 · Cited by 31,241 · cited 3×
“The selection criterion can be applied systematically to diagrams of any size and shape, thus freeing analysts from judging whether “X is conditionally ignorable given S,” a formidable mental task required in the potential-response framework (Rosenbaum and Rubin, 1983).”
BackgroundMethod(2001) · RePEc: Research Papers in Economics · Cited by 27,577
Structural Equations with Latent Variables.
Clogg, C. C., & Bollen, K. A. (1991) · Contemporary Sociology · doi:10.2307/2072165 · Cited by 17,442
Structural Equations with Latent Variables
Bollen, K. A. (1989) · Wiley series in probability and statistics · doi:10.1002/9781118619179 · Cited by 17,277 · cited 3×
“…identification has received considerable attention in econometrics (Hurwicz, 1950; Marschak, 1950; Koopmans, 1953) and social science (Duncan, 1975; Bollen, 1989), usually in linear parametric settings, were it reduces to asking whether some model parameter, β, has a unique solution in terms of…”
Causality: Models, Reasoning, and Inference
Pearl, J. (2009) · doi:10.1017/cbo9780511803161 · Cited by 11,653
Self-citationMediation in experimental and nonexperimental studies: New procedures and recommendations.
Shrout, P. E., & Bolger, N. (2002) · Psychological Methods · doi:10.1037/1082-989x.7.4.422 · Cited by 11,335 · cited 2×
“…) = ∑ z E(Y |x, z)[P (z|x′)− P (z|x)] (52) This expression deserves the label Mediation Formula, due to its pivotal role in mediation analysis (Imai et al., 2008), which has been a thorny issue in several sciences (Shrout and Bolger, 2002; MacKinnon et al., 2007; Mortensen et al., 2009).”
BackgroundEstimating causal effects of treatments in randomized and nonrandomized studies
Rubin, D. B. (1974) · Journal of Educational Psychology · doi:10.1037/h0037350 · Cited by 9,436 · cited 4×
“…equation models (SEM) used in economics and social science (Goldberger, 1973; Duncan, 1975), the potentialoutcome framework of Neyman (1923) and Rubin (1974), and the graphical models developed for probabilistic reasoning and causal analysis (Pearl, 1988; Lauritzen, 1996; Spirtes et al., 2000;…”
BackgroundMethodInfluentialMacKinnon, D. P., Fairchild, A. J., & Fritz, M. S. (2007) · Annual Review of Psychology · doi:10.1146/annurev.psych.58.110405.085542 · Cited by 5,688 · cited 2×
“…) = ∑ z E(Y |x, z)[P (z|x′)− P (z|x)] (52) This expression deserves the label Mediation Formula, due to its pivotal role in mediation analysis (Imai et al., 2008), which has been a thorny issue in several sciences (Shrout and Bolger, 2002; MacKinnon et al., 2007; Mortensen et al., 2009).”
BackgroundDiscrete Multivariate Analysis: Theory and Practice.
Beniger, J. R., Bishop, Y. M. M., Feinberg, S. E., Holland, P. W., Light, R. J., & Mosteller, F. (1975) · Contemporary Sociology · doi:10.2307/2063625 · Cited by 5,108
Recent Developments in the Econometrics of Program Evaluation
Imbens, G. W., & Wooldridge, J. M. (2009) · Journal of Economic Literature · doi:10.1257/jel.47.1.5 · Cited by 4,926
Research works printed but not linked: 11
- Glymour, M. and Greenland, S. (2008). Causal diagrams. In, Modern Epidemiology (K. Rothman, S. Greenland and T. Lash, eds.), 3rd ed. Lippincott Williams & Wilkins, Philadelphia, PA, 183–209.
- Hurwicz, L. (1950). Generalization of the concept of identification. In, Statistical Inference in Dynamic Economic Models (T. Koopmans, ed.). Cowles Commission, Monograph 10, Wiley, New York, 245–257.
- Koopmans, T. (1953). Identification problems in econometric model construction. In, Studies in Econometric Method (W. Hood and T. Koopmans, eds.). Wiley, New York, 27–48.
- Marschak, J. (1950). Statistical inference in economics. In, Statistical Inference in Dynamic Economic Models (T. Koopmans, ed.). Wiley, New York, 1–50. Cowles Commission for Research in Economics, Monograph 10.
- Pearl, J. (1988)., Probabilistic Reasoning in Intelligent Systems. Morgan Kaufmann, San Mateo, CA.
- Pearl, J. (2001). Direct and indirect effects. In, Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence. Morgan Kaufmann, San Francisco, CA, 411–420.
- Robins, J. (1989). The analysis of randomized and non-randomized aids treatment trials using a new approach to causal inference in longitudinal studies. In, Health Service Research Methodology: A Focus on AIDS (L. Sechrest, H. Freeman and A. Mulley, eds.). NCHSR, U.S. Public Health Service, Washington, D.C., 113–159.
- Wermuth, N. (1992). On block-recursive regression equations., Brazilian Journal of Probability and Statistics (with discussion) 6 1–56.
- Cole, P. (1997). Causality in epidemiology, health policy, and law., Journal of Marketing Research 27 10279–10285.
- Pearson, K., Lee, A. and Bramley-Moore, L. (1899). Genetic (reproductive) selection: Inheritance of fertility in man., Philosophical Transactions of the Royal Society A 73 534–539.
- Robertson, D. (1997). The common sense of cause in fact., Texas Law Review 75 1765–1800.
Cited by
Items 601–700 of 2,354
RETRACTED: Robust policy evaluation from large-scale observational studies
Islam, M. S., Morshed, M. S., Young, G. J., & Noor‐E‐Alam, M. (2019) · PLoS ONE · doi:10.1371/journal.pone.0223360 · Cited by 20
“The impact assessment is not associational but causal [1, 2] which requires an understanding of the counterfactual—the difference in outcomes with or without the presence of the policy [3].”
RetractedGenerative Counterfactuals for Neural Networks via Attribute-Informed Perturbation
Yang, F., Liu, N., Du, M., & Hu, X. (2021) · ACM SIGKDD Explorations Newsletter · doi:10.1145/3468507.3468517 · Cited by 32
“However, existing correlation-based explanations are neither discriminative nor counterfactual [29], since they are not able to help understand why and how particular explanations are relevant to model decisions.”
BackgroundWang, X., Lee, Y., Zhu, X., & Okur, A. (2021) · Research in Higher Education · doi:10.1007/s11162-021-09644-w · Cited by 27
“The ignorability assumption cannot be fully verified given the unobserved nature of such a confounder (Pearl, 2009).”
BackgroundExploratory Causal Analysis with Time Series Data
McCracken, J. M. (2016) · Synthesis lectures on data mining and knowledge discovery · doi:10.2200/s00707ed1v01y201602dmk012 · Cited by 22
Rough Sets Turn 40: From Information Systems to Intelligent Systems
Skowron, A., & Ślȩzak, D. (2022) · Annals of Computer Science and Information Systems · doi:10.15439/2022f310 · Cited by 22
“This fits the paradigm of computing with words [117] (which also corresponds to the foundations of information granulation with respect to decomposing complex problems onto their smaller components) and, in particular, the following challenge formulated by Judea Pearl [118]: Traditional statistics is strong in devising…”
BackgroundGorade, V., Mittal, S., & Singhal, R. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.107569 · Cited by 20
Computational Experiments: A New Analysis Method for Cyber-Physical-Social Systems
Xue, X., Shen, Y., Yu, X., et al. (2023) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2023.3322402 · Cited by 17
“To further quantify the causal relationship between the two nodes, the probabilistic causal model (PCM) [26] can be used to formalize and define relevant operations.”
MethodToward identifying malicious encrypted traffic with a causality detection system
Zeng, Z., Peng, X., Peng, W., & Zhao, B. (2023) · Journal of Information Security and Applications · doi:10.1016/j.jisa.2023.103644 · Cited by 22
Chen, X., Sun, R., Saluz, U., Schiavon, S., & Geyer, P. (2023) · Developments in the Built Environment · doi:10.1016/j.dibe.2023.100296 · Cited by 27
Regression analyses and their particularities in observational studies
Zapf, A., Wiessner, C., & König, I. R. (2024) · Deutsches Ärzteblatt international · doi:10.3238/arztebl.m2023.0278 · Cited by 36
Ten questions regarding buildings, occupants and the energy performance gap
Mahdavi, A., & Berger, C. (2024) · Journal of Building Performance Simulation · doi:10.1080/19401493.2024.2332245 · Cited by 17
“In such cases, it may be possible to explore causal relationship via retrospective studies and causal inference approaches (Nogueira et al. 2022; Pearl 2009).”
Explainable AI for Earth observation: current methods, open challenges, and opportunities
Taşkın, G., Aptoula, E., & Ertürk, A. (2024) · Elsevier eBooks · doi:10.1016/b978-0-44-319077-3.00012-2 · Cited by 24
Dynamic Information Retrieval Modeling
Yang, G. H., Sloan, M., & Wang, Y.-X. (2016) · Synthesis lectures on information concepts, retrieval, and services · doi:10.2200/s00718ed1v01y201605icr049 · Cited by 23
Uhde, C., Berberich, N., Ramírez-Amaro, K., & Cheng, G. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9341505 · Cited by 24
“It has been shown that causal structures can be inferred through interventions and controlled experiments [15][16].”
BackgroundOn Hens, Eggs, Temperatures and CO2: Causal Links in Earth’s Atmosphere
Koutsoyiannis, D., Onof, C., Kundzewicz, Z. W., & Christofides, A. (2023) · Sci · doi:10.3390/sci5030035 · Cited by 26
Nonparametric Identification of Causal Effects Under Temporal Dependence
Dafoe, A. (2014) · Sociological Methods & Research · doi:10.1177/0049124115613784 · Cited by 22
Review article: Hilbert problems for the climate sciences in the 21st century – 20 years later
Ghil, M. (2020) · Nonlinear processes in geophysics · doi:10.5194/npg-27-429-2020 · Cited by 19
“…well being (e.g., Ghil et al., 2011; Chavez et al., 2015; Lucarini et al., 2016), an important step in achieving greater rigor in this field is a greater reliance on the counter-factual theory of necessary and sufficient causation due to J. Pearl (Pearl, 2009a, b) in the attribution of such events.”
Jiang, F., & Kazmi, H. (2024) · Applied Energy · doi:10.1016/j.apenergy.2024.124550 · Cited by 24
Massively Parallel Causal Inference of Whole Brain Dynamics at Single Neuron Resolution
Watanakeesuntorn, W., Takahashi, K., Ichikawa, K., Park, J. C., Sugihara, G., Takano, R., Haga, J., & Pao, G. M. (2020) · IEEE International Conference on Parallel and Distributed Systems (ICPADS) · doi:10.1109/icpads51040.2020.00035 · Cited by 20
“Structural Causal Model (SCM) is one of the most popular causal models [12] based on statistical modeling of equilibrium systems.”
BackgroundMunson, M. R., Raghavan, R., Shimizu, R., Rodwin, A. H., & Jaccard, J. J. (2022) · Psychiatric Services · doi:10.1176/appi.ps.202100202 · Cited by 16
Counterfactual inference graph network for disease prediction
Zhang, B., Guo, X., Lin, Q., Wang, H., & Songbai, X. (2022) · Knowledge-Based Systems · doi:10.1016/j.knosys.2022.109722 · Cited by 18
Naser, M. Z., & Çiftçioğlu, A. Ö. (2023) · Structural Concrete · doi:10.1002/suco.202200525 · Cited by 16
“A broader look into causality can be found herein.(43,64,65) † The reader is highly encouraged to visit the following sources on causality, do-calculus, and others.”
BackgroundA causal inference method for improving the design and interpretation of safety research
Niu, Y., Fan, Y., Gao, Y., & Li, Y. (2023) · Safety Science · doi:10.1016/j.ssci.2023.106082 · Cited by 16
Gabler, P., Geiger, B. C., Schuppler, B., & Kern, R. (2023) · Information · doi:10.3390/info14020137 · Cited by 34
“Causal statements requiring such assumptions are often eschewed, be it on epistemological grounds or simply due to a lack of mathematical formalization, but can in fact be made statistically rigorous [62].”
Learning Causal Representations for Generalizable Face Anti Spoofing
Zheng, G., Liu, Y., Dai, W., Li, C., Zou, J., & Xiong, H. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10095329 · Cited by 17
“To this end, we propose a novel causal method, namely CRFAS, that models the data generation process of FAS via SCM [14] and extracts causal features by applying backdoor adjustment [14] to exclude the impact of non-causal features.”
BackgroundMethodInfluentialNunney, E., Crotta, M., Winden, S. C. L. V., Bond, K., Green, M. J., & Guitián, J. (2023) · Preventive Veterinary Medicine · doi:10.1016/j.prevetmed.2023.106022 · Cited by 15
Priority Challenges for Social and Behavioral Research and Its Modeling
Davis, P. K., O’Mahony, A., & Gulden, T. R. (2017) · RAND Corporation eBooks · doi:10.7249/wr1206 · Cited by 16
Lin, S., Huang, Y., & Yang, H. (2019) · Statistics in Medicine · doi:10.1002/sim.8123 · Cited by 16
Beyerer, J., Maier, A., & Niggemann, O. (2020) · Technologien für die intelligente Automation · doi:10.1007/978-3-662-62746-4 · Cited by 17
A Consumer Behavior Prediction Model Based on Multivariate Real-Time Sequence Analysis
Guo, L., Zhang, B., & Zhao, X. (2021) · Mathematical Problems in Engineering · doi:10.1155/2021/6688750 · Cited by 19
Causal inference in regression: advice to authors
Bulbulia, J. A., Schjoedt, U., Shaver, J. H., Sosis, R. H., & Wildman, W. J. (2021) · Religion Brain & Behavior · doi:10.1080/2153599x.2021.2001259 · Cited by 15
“We recommend that researchers use directed acyclic graphs (DAGs) to systematically identify and avoid collider (and other) forms of confounding (McElreath, 2020; Pearl, 2009; Pearl, 2019).”
MethodUnderstanding Overlap in Automatic Root Cause Analysis in Manufacturing Using Causal Inference
Oliveira, E. E., Miguéis, V. L., & Borges, J. L. (2021) · IEEE Access · doi:10.1109/access.2021.3139199 · Cited by 16
“for the analysis of causes and counterfactuals [17].”
BackgroundIntrusion detection framework based on causal reasoning for DDoS
Zeng, Z., Peng, W., Zeng, D., Zeng, C. Y., & Chen, Y. (2022) · Journal of Information Security and Applications · doi:10.1016/j.jisa.2022.103124 · Cited by 22
Reversals of Least-Square Estimates and Model-Invariant Estimation for Directions of Unique Effects
Knaeble, B., & Dutter, S. (2016) · The American Statistician · doi:10.1080/00031305.2016.1226951 · Cited by 25
“Ideally, the covariates W1, . . . ,Wp have been selected so that conditional on W = (W1, . . . ,Wp) the remaining variation in X is caused by factors that are independent of Y , although this nonconfounding condition may be difficult to verify (Pearl 2009b, p. 350).”
BackgroundInfluentialRacial disparities in infant mortality: what has birth weight got to do with it and how large is it?
Gage, T. B., Fang, F., O’Neill, E., & DiRienzo, A. G. (2010) · BMC Pregnancy and Childbirth · doi:10.1186/1471-2393-10-86 · Cited by 16
“The result is called a “generated direct effect” by Geneletti [18], which is similar to Pearl’s “natural direct effect” [24].”
MethodResultLombardi, C. M., & Coley, R. L. (2016) · Child Development · doi:10.1111/cdev.12588 · Cited by 16
Kaiser, P., Arnold, A. M., Benkeser, D. C., Hazzouri, A. Z. A., Hirsch, C. H., Psaty, B. M., & Odden, M. C. (2017) · International Journal of Epidemiology · doi:10.1093/ije/dyx179 · Cited by 17
Zhao, S., & Heffernan, N. T. (2017) · Educational Data Mining · Cited by 18
Modeling Social Networks as Mediators: A Mixed Membership Stochastic Blockmodel for Mediation
Sweet, T. M. (2018) · Journal of Educational and Behavioral Statistics · doi:10.3102/1076998618814255 · Cited by 22
“For readers interested in more rigorous statistical methods for estimating causal inference, we suggest Holland (1986) or Rubin (2005) for the potential outcomes framework and Pearl (2009) for the directed-acyclic graph perspective.”
MethodThe Mathematics of Causal Relations
Pearl, J. (2011) · Oxford University Press eBooks · doi:10.1093/oso/9780199754649.003.0007 · Cited by 18
Self-citationDual Instrumental Variable RegressionPreprint
Muandet, K., Mehrjou, A., Lee, S. K., & Raj, A. (2019) · RePEc: Research Papers in Economics · doi:10.48550/arxiv.1910.12358 · Cited by 28
"Telling a Story." On the Dramaturgy of Monophonic Jazz Solos
Frieler, K., Pfleiderer, M., Abe, J., et al. (2016) · Empirical Musicology Review · doi:10.18061/emr.v11i1.4959 · Cited by 15
“Initially, we extracted a slightly larger set of features for each phrase, but due to the inherent correlations (by mathematical construction), we pruned the set of features using causal inference (Pearl, 2009) as an informal way to identify the most fundamental features.”
Answerable and Unanswerable Questions in Risk Analysis with Open‐World Novelty
Cox, L. A. (2020) · Risk Analysis · doi:10.1111/risa.13553 · Cited by 19
“…proved highly useful for identifying and quantifying the causal roles of confounders, modifiers, and mediators in epidemiological risk assessment (Pearl 2009) and social science applications, where population responses to changes in causal factors may unfold on a time scale of years to decades,…”
BackgroundPelham, W. E., Tapert, S. F., Gonzalez, M. R., Guillaumé, M., Dick, A. S., Sheth, C. S., Baker, F. C., Baskin-sommers, A. R., Marshall, A. T., Lisdahl, K. M., et al. (2022) · Research on Child and Adolescent Psychopathology · doi:10.1007/s10802-021-00896-9 · Cited by 18
“Because it cannot be verified that a given variable is truly a confounder (Miller & Chapman, 2001; Pearl, 2009), we took the approach of repeating each analysis while including/ excluding different sets of covariates in order to determine to what extent conclusions hinged on the specific covariates included.”
MethodLiu, Q., Luo, Y., Wu, S., Zhang, Z., Yue, X., Jin, H., & Wang, L. (2022) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2022.3179025 · Cited by 17
Introducing Causal Inference Using Bayesian Networks and do-Calculus
Lu, Y., Zheng, Q., & Quinn, D. (2022) · Journal of Statistics and Data Science Education · doi:10.1080/26939169.2022.2128118 · Cited by 15
Explaining Technology We Do Not Understand
Adamson, G. (2023) · IEEE Transactions on Technology and Society · doi:10.1109/tts.2023.3240107 · Cited by 18
“Following causal analysis we can go further, to infer results under changing conditions [81].”
Income raises human well-being indefinitely, but age consistently slashes it
Li, C., & Managi, S. (2023) · Scientific Reports · doi:10.1038/s41598-023-33235-7 · Cited by 21
A Perron–Frobenius analysis of wall-bounded turbulence
Jiménez, J. (2023) · Journal of Fluid Mechanics · doi:10.1017/jfm.2023.515 · Cited by 15
Taylor, J., & Kriegeskorte, N. (2023) · Scientific Reports · doi:10.1038/s41598-023-40807-0 · Cited by 24
Shinn, L. M., Mansharamani, A., Baer, D. J., Novotny, J. A., Charron, C. S., Khan, N. A., Zhu, R., & Holscher, H. D. (2023) · Journal of Nutrition · doi:10.1016/j.tjnut.2023.11.001 · Cited by 18
On Causality in Distributed Continuum Systems
Pujol, V. C., Sedlak, B., Donta, P. K., & Dustdar, S. (2024) · IEEE Internet Computing · doi:10.1109/mic.2023.3344248 · Cited by 22
“Currently, there are two main frameworks for modeling causality: structural causal models (SCMs), 16 based on causal graphs and structural equations, and the POs framework, which inherits from the development of randomized experiments (see Yao et al. 17 for a comprehensive survey).”
BackgroundA survey on cell nuclei instance segmentation and classification: Leveraging context and attention
Nunes, J. D., Montezuma, D., Oliveira, D., Pereira, T., & Cardoso, J. S. (2024) · Medical Image Analysis · doi:10.1016/j.media.2024.103360 · Cited by 23
Perspectives on Uncertainty and Risk in Rough Sets and Interactive Rough-Granular Computing
Jankowski, A. W., Skowron, A., & Świniarski, R. W. (2014) · Fundamenta Informaticae · doi:10.3233/fi-2014-961 · Cited by 15
“Following Judea Pearl [31], one can say that traditional statistical approach to risk management inference is strong in devising ways of describing data and inferring distributional parameters from sample.”
Hoffman, S., & Podgurski, A. (2011) · The Journal of Law Medicine & Ethics · doi:10.1111/j.1748-720x.2011.00612.x · Cited by 20
Bry, X., Robette, N., & Roueff, O. (2015) · Quality & Quantity · doi:10.1007/s11135-015-0187-z · Cited by 15
“An interesting approach is developed by Judea Pearl in the framework of analytical sociology, based on structural causality (Pearl 2009).”
MethodHealth Services Utilization in Older Europeans: an Empirical Study
Srakar, A., Hren, R., & Rupel, V. P. (2016) · Organizacija · doi:10.1515/orga-2016-0009 · Cited by 15
Conditions for handling confounding variables in dynamic networks
Dankers, A. G., den Hof, P. M. J. V., Materassi, D., et al. (2017) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2017.08.771 · Cited by 15
“In the probabilistic inference literature, a confounding variable is defined as an unmeasured variable that causally affects both wi and wj [Pearl, 2009].”
BackgroundLiving with observational data in biological anthropology
Smith, R. J. (2019) · American Journal of Physical Anthropology · doi:10.1002/ajpa.23862 · Cited by 16
Vanhatalo, J., Hartmann, M., & Veneranta, L. (2019) · Bayesian Analysis · doi:10.1214/19-ba1158 · Cited by 15
“be naturally tackled with predictive causal inference (Lindley, 2002; Pearl, 2009).”
BackgroundA Bridge from Monty Hall to the Hot Hand: The Principle of Restricted Choice
Miller, J. B., & Sanjurjo, A. (2019) · The Journal of Economic Perspectives · doi:10.1257/jep.33.3.144 · Cited by 15
Information Theoretic Causal Effect Quantification
Wieczorek, A., & Röth, V. (2019) · Entropy · doi:10.3390/e21100975 · Cited by 21
Belbahri, M., Murua, A., Gandouet, O., & Nia, V. P. (2021) · The Annals of Applied Statistics · doi:10.1214/21-aoas1465 · Cited by 17
Zepeda, J. A., Rödel, H. G., Monclús, R., Hudson, R., & Bautista, A. (2019) · Behavioral Ecology and Sociobiology · doi:10.1007/s00265-019-2777-6 · Cited by 16
“We applied d-separation to the set of independence claims, and for that, we fitted linear mixed-effects models with a Gaussian distribution, and with litter identity as a random factor (Pearl 2009; Shipley 2009).”
MethodOptimizing Variational Graph Autoencoder for Community Detection
Choong, J. J., Liu, X., & Murata, T. (2019) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata47090.2019.9006123 · Cited by 14
“Causal inference [9] and Bayesian Deep Learning [10] are some examples of teaching machine learning algorithms to describe uncertainties and causal reasons.”
BackgroundWang, Z., Chapman, D., Morota, G., & Cheng, H. (2020) · G3 Genes Genomes Genetics · doi:10.1534/g3.120.401618 · Cited by 14
“GWAS have successfully mapped quantitative trait loci (QTL) associated with traits of interest, e.g. , meat quality and quantity in livestock (Sharma et al. 2015), crop yields in plants (Liu and Yan 2018), and diseases in humans (Visscher et al. 2017).”
BackgroundHülsenbeck, I., Frank, M., Biewald, E. M., Kanber, D., Lohmann, D. R., & Ketteler, P. (2021) · Cancers · doi:10.3390/cancers13071605 · Cited by 13
“In other words, pRB is the last known mediator toward phenotypic expression (see [41] for a short introduction to causal inference).”
BackgroundPaths academic scientists take to entrepreneurship: Disaggregating direct and indirect influences
Dohse, D., Goel, R. K., & Göktepe‐Hultén, D. (2021) · Managerial and Decision Economics · doi:10.1002/mde.3341 · Cited by 20
“These indirect influences are not evident without the mediation analysis (see Baron & Kenny, 1986; Goel, 2020; Pearl, 2009) and open additional policy avenues to facilitate academic entrepreneurship.”
BackgroundMethodHarmonization with Flow-Based Causal Inference
Wang, R., Chaudhari, P., & Davatzikos, C. A. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-87199-4_17 · Cited by 23
Promise and peril of ecological and evolutionary modelling using cross-cultural datasets
Bird, R. B., & Codding, B. F. (2021) · Nature Ecology & Evolution · doi:10.1038/s41559-021-01579-w · Cited by 16
Uehleke, R., Petrick, M., & Hüttel, S. (2022) · Land Use Policy · doi:10.1016/j.landusepol.2021.105950 · Cited by 23
HypeR: Hypothetical Reasoning With What-If and How-To Queries Using a Probabilistic Causal Approach
Galhotra, S., Gilad, A., Roy, S., & Salimi, B. (2022) · ACM SIGMOD International Conference on Management of Data (SIGMOD) · doi:10.1145/3514221.3526149 · Cited by 21
“To do so, we leverage the notion of backdoor criterion from causal inference [38].”
BackgroundMethodInfluentialCausal analysis, Correlation-Response, and Dynamic cavity
Aurell, E., & Ferraro, G. D. (2016) · Journal of Physics Conference Series · doi:10.1088/1742-6596/699/1/012002 · Cited by 17
“Nevertheless, causal analysis is an important branch of statistics, describing the effects of interventions and answering questions of the “if-so-then-what?” character [8, 9, 10].”
BackgroundCosta, A. B. P., Machado, L. A. C., Telles, R. W., & Barreto, S. M. (2020) · Rheumatology International · doi:10.1007/s00296-020-04557-w · Cited by 17
“Additionally, the 2-step adjustment procedure used in our analysis allowed judgmental assumptions regarding causal relationships to assist the selection of covariates for the final regression models (theory-driven approach), also reducing the risk of confounding [60]; e.”
MethodInferring network structure from interventional time-course experiments
Spencer, S. E. F., Hill, S. M. (., & Mukherjee, S. (2015) · Warwick Research Archive Portal (University of Warwick) · Cited by 15
Improving the Accuracy of Network Intrusion Detection with Causal Machine Learning
Zeng, Z., Peng, W., & Zhao, B. (2021) · Security and Communication Networks · doi:10.1155/2021/8986243 · Cited by 22
Chetia, T., Baruah, S., Dey, C., Baruah, S., & Sharma, S. (2022) · Natural Hazards · doi:10.1007/s11069-021-05168-9 · Cited by 15
“The correlation observed was moderate to strong with r ~ 0.63 with a high statistical significance of p 0.0001 but it does not imply the cause and effect relationship (Pearl 2009, 2010).”
BackgroundInfluentialDefinition and Diagnosis of Problematic Attrition in Randomized Controlled ExperimentsPreprint
García, F. M. (2013) · SSRN Electronic Journal · doi:10.2139/ssrn.2302735 · Cited by 13
Sage: Leveraging ML to Diagnose Unpredictable Performance in Cloud Microservices
Gan, Y., Liang, M., Dev, S., Lo, D., & Delimitrou, C. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2112.06263 · Cited by 15
“To avoid impacting the performance of live services during this process, Sage leverages historical tracing data to generate realistic counterfactuals [41, 46], taking into account that the exact situation may not have occurred in the past.”
BackgroundAn ontology-based text mining dataset for extraction of process-structure-property entities
Durmaz, A. R., Thomas, A., Mishra, L., Murthy, R. N., & Straub, T. (2024) · Scientific Data · doi:10.1038/s41597-024-03926-5 · Cited by 27
Transformation priming helps to disambiguate sudden changes of sensory inputs
Pastukhov, A., Vivian-Griffiths, S., & Braun, J. (2015) · Vision Research · doi:10.1016/j.visres.2015.09.005 · Cited by 11
“To this end, we compared conditional probabilities for pairs of trials with same history of transformations (stability or change) but opposite history of perceptual states, to test a causal relationship between perception of states and perception of transformations (Pearl, 2009), see Fig.”
BackgroundUnintentional Action Localization via Counterfactual Examples
Xu, J., Chen, G., Lu, J., & Zhou, J. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3166278 · Cited by 14
Betkó, J., Spierings, N., Gesthuizen, M., & Scheepers, P. L. H. (2022) · Basic Income Studies · doi:10.1515/bis-2021-0029 · Cited by 20
Zhang, Z., Akinci, B., & Qian, S. (2023) · Accident Analysis & Prevention · doi:10.1016/j.aap.2023.106966 · Cited by 19
Interpretability for reliable, efficient, and self-cognitive DNNs: From theories to applications
Kang, X., Guo, J., Song, B., Cai, B., Sun, H., & Zhang, Z. (2023) · Neurocomputing · doi:10.1016/j.neucom.2023.126267 · Cited by 16
Disentangled Representation for Causal Mediation Analysis
Xu, Z., Cheng, D., Li, J., Liu, J., Liu, L., & Wang, K. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i9.26266 · Cited by 31
“…(Pearl 2014), the counterfactual expression of NDE (Equation 2) and NIE r (Equation 5) can be reduced to do-expressions as follows: Following Pearl’s back-door adjustment formula (Pearl 2009a,b), we propose the following theorem which is used to reduce the do-expressions to probability expressions.”
DCMT: A Direct Entire-Space Causal Multi-Task Framework for Post-Click Conversion Estimation
Zhu, F., Zhong, M., Yang, X., Li, L., Lu, Y., Zhang, T., Zhou, J., Chen, C., Wu, F. Q., Liu, G., et al. (2023) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde55515.2023.00239 · Cited by 46
“Causal inference is to study the independent and actual effect of a particular component in a system [34].”
BackgroundFuria, C. A., Torkar, R., & Feldt, R. (2024) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3611667 · Cited by 14
“1 Causal Inference Causal inference denotes a process, usually built on top of mathematical/statistical concepts, for identifying and analyzing which factors lead to specific outcomes [40, 44].”
BackgroundEvans, L. C., Burgess, M. D., Potts, S. G., Kunin, W. E., & Oliver, T. H. (2024) · Ecology Letters · doi:10.1111/ele.14362 · Cited by 18
Causal relationships in longitudinal observational data: An integrative modeling approach.
Biazoli, C. E., Sato, J. R., & Pluess, M. (2024) · Psychological Methods · doi:10.1037/met0000648 · Cited by 16
Using neural biomarkers to personalize dosing of vagus nerve stimulation
Berthon, A., Wernisch, L., Stoukidi, M., Thornton, M., Tessier-Lariviere, O., Fortier-Poisson, P., Mamen, J., Pinkney, M., Lee, S., Sarkans, E., et al. (2024) · Bioelectronic Medicine · doi:10.1186/s42234-024-00147-4 · Cited by 18
Toward Intelligent Attack Detection With Causal Transformer in Internet of Things
Zeng, Z., Zhao, B., Deng, X., Liu, X., Zheng, J., & Chen, J. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3512531 · Cited by 16
“In addition, when using the correlation method to identify causality, machine learning may lead to false conclusions in all cases except in the simplest causal situations [12].”
Causal Inference from Big Data: Theoretical Foundations and the Data-fusion Problem
Bareinboim, E., & Pearl, J. (2015) · Cited by 15
Self-citationMining structural influence to analyze relationships in social network
Guo, L., & Zhang, B. (2019) · Physica A Statistical Mechanics and its Applications · doi:10.1016/j.physa.2019.02.005 · Cited by 14
Ferrari, E., Retico, A., & Bacciu, D. (2020) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2020.101804 · Cited by 22
“Previous renowned works on this subject are the ’Back-door’ and ’Frontdoor’ criteria, developed in the causal inference framework described in Judea Pearl’s work [23, 24], commonly cited as a way to determine which variables play as counfounders.”
BackgroundKheyfets, V. O., Dufva, M. J., Boehm, M., Tian, X., Qin, X., Tabakh, J. E., Truong, U. T., Ivy, D. D., & Spiekerkoetter, E. F. (2020) · Physiological Reports · doi:10.14814/phy2.14347 · Cited by 14
“Future studies will be prospectively designed to draw causal inference (Pearl, 2009) between RV mechanics, LV mechanics, and gene expression.”
BackgroundPrier, E., McCue, C. P., & Csáki, C. (2020) · International Journal of Public Administration · doi:10.1080/01900692.2020.1773497 · Cited by 15
“For example, the correlational analysis here does not evaluate causality that might be done through tools like directed acyclic graphs (see Pearl, 2009), nor does it consider other potential confounding factors that might indicate spurious or causal associations.”
BackgroundMeloncelli, N. J. L., Barnett, A. G., & de Jersey, S. (2020) · BMC Pregnancy and Childbirth · doi:10.1186/s12884-020-03352-6 · Cited by 17
Supporting Mental Health at Work (Comment on “The Epidemic of Mental Disorders in Business”)
Pierce, L., & Rider, C. I. (2022) · Administrative Science Quarterly · doi:10.1177/00018392211072479 · Cited by 18
“Distinguishing harmful contagion from other plausible mechanisms (including beneficial contagion) is, however, central to establishing the causal relationship that KAL theorize (Pearl, 2009; Hedström and Ylikoski, 2010).”
BackgroundExploring uplift modeling with high class imbalance
Nyberg, O., & Arto Klami, A. (2023) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-023-00917-9 · Cited by 18
Metadata
- Authors
- Pearl, J.
- Affiliations
- University of California, Los Angeles (US)
- Corresponding author
- Pearl, Judea
- Venue
- Statistics Surveys
- Publisher
- Institute of Mathematical Statistics
- Published
- 2009-01-01
- Volume
- 3
- Issue
- none
- Pages
- 96-146
- ISSN
- 1935-7516 (ISSN-L)
- OpenAlex
- W2143117649
- Cited by
- 2,336
- References
- 135
- Citations by year
- 2026: 184 · 2025: 273 · 2024: 271 · 2023: 245 · 2022: 215 · 2021: 208 · 2020: 243 · 2019: 140 · 2018: 96 · 2017: 77 · 2016: 82 · 2015: 82 · 2014: 63 · 2013: 65 · 2012: 46
- FWCI
- 25.47
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Statistics Surveys (publishedVersion, cc-by)
- Project Euclid (Cornell University) (submittedVersion)
- eScholarship (California Digital Library) (submittedVersion)
- Department of Statistics, UCLA
- Funders
- Office of Naval Research
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Bayesian Modeling and Causal Inference
- Social Sciences › Arts and Humanities › History and Philosophy of Science › Philosophy of Biology and Scientific Explanation
- Keywords
- causal inference, structural causal models, counterfactual reasoning, causal assumptions, causal effects, policy evaluation, intervention effects, regret, causal attribution, direct and indirect effects, mediation analysis
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.