pith. sign in

arxiv: 2604.09153 · v1 · submitted 2026-04-10 · 💻 cs.CR

Hagenberg Risk Management Process (Part 3): Operationalization, Probabilities, and Causal Analysis

Eckehard Hermann , Harald Lampesberger This is my paper

Pith reviewed 2026-05-10 17:33 UTC · model grok-4.3

classification 💻 cs.CR
keywords risk managementbowtie diagramsbayesian inferencecausal analysisdirected acyclic graphsprobabilistic modelsexpert elicitation
0
0 comments X

The pith

Bowtie risk diagrams convert into probabilistic DAGs that support Bayesian inference and causal interventions.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes a method to operationalize risk management for hazards that cannot be fully eliminated by extending qualitative Bowtie diagrams into directed acyclic graphs. These graphs incorporate safe-state definitions and activation nodes to enable real-time probabilistic modeling and Bayesian analysis. Tools capture expert probability estimates, handle disagreements through dispersion analysis and regularization, and apply causal techniques such as d-separation and do-calculus for what-if scenarios. A sympathetic reader would care because the approach turns static risk assessments into observable, intervention-ready models illustrated on an instant-payments gateway example.

Core claim

The Hagenberg Risk Management Process transforms Bowtie structures of causes, top events, barriers, and consequences into a DAG suitable for Bayesian inference. Realtime Risk Studio adds explicit safe-state semantics and designates Activation Nodes as intervention points. Probability Capture generates expert questionnaires, aggregates conditional probabilities despite disagreement using descriptive and prior-regularized methods, and enables causal analysis via d-separation, adjustment sets, do-calculus, and impact searches. The workflow is shown on a payments gateway without statistical validation, positioning the contribution as a method and prototype system.

What carries the argument

Bowtie-to-DAG transformation in Realtime Risk Studio, which converts qualitative structures into Bayesian networks while preserving risk elements and adding safe-state and activation-node semantics.

If this is right

  • Risks that cannot be accepted or mitigated become continuously observable through runtime probabilistic models.
  • Expert disagreements on probabilities become quantifiable and manageable via dispersion analysis and regularization.
  • Causal what-if analysis using do-calculus identifies effective interventions at Activation Nodes.
  • Models distinguish safe states explicitly from risk states for clearer operational decisions.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The approach could support live data integration to update probabilities automatically beyond expert input alone.
  • It opens pathways for combining the method with observed incident data to refine causal estimates over time.
  • Similar transformations might apply to Bowtie use in safety-critical engineering fields outside cybersecurity.

Load-bearing premise

Qualitative Bowtie structures can be turned into DAGs that retain essential risk meaning for Bayesian use, and expert probability estimates can be aggregated reliably even when experts disagree.

What would settle it

Apply the full workflow to a risk scenario with independently measured outcome probabilities and check whether the model's Bayesian inferences and recommended interventions align with the observed data.

Figures

Figures reproduced from arXiv: 2604.09153 by Eckehard Hermann, Harald Lampesberger.

Figure 1
Figure 1. Figure 1: Bowtie-derived runtime DAG of the instant-payments risk in the Realtime Risk Studio. [PITH_FULL_IMAGE:figures/full_fig_p013_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: DAG view and CPT editor for the node Service Degradation Under Load. Table II ILLUSTRATIVE CONSEQUENCE-STATE POSTERIORS FOR THREE OPERATIONAL SITUATION CUTS IN THE INSTANT-PAYMENTS EXAMPLE. Cut Evidence Safe Degr. Part. Loss 1 Peak load, normal queue, low latency, rollback available 0.94 0.05 0.01 0.00 2 Faulty change, elevated queue, high latency, retry state unclear 0.46 0.38 0.13 0.03 3 Critical queue, … view at source ↗
Figure 3
Figure 3. Figure 3: Noise-analysis view for conditional probability questions in the instant-payments example. [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Causal-analysis view for the instant-payments case. [PITH_FULL_IMAGE:figures/full_fig_p016_4.png] view at source ↗
read the original abstract

For risks that cannot be accepted, sufficiently mitigated, or eliminated, continuous observation is a viable approach but requires a model that can be operationalized. The Hagenberg Risk Management Process bridges this gap between qualitative risk analysis, using contextualized polar heatmaps (triage), and realtime risk management by extending Bowtie diagrams into a formal probabilistic runtime model. We introduce Realtime Risk Studio, a domain-specific modeling tool that (i) transforms Bowtie structures (causes, top event, barriers, consequences) into a directed acyclic graph (DAG) suitable for Bayesian inference, (ii) adds explicit safe-state semantics, and (iii) designates Activation Nodes as intervention points. Bowtie models are qualitative; however, Bayesian inference requires actual probabilities. As a second contribution, we present Probability Capture, a tool that complements our Realtime Risk Studio by automatically generating questionnaires from a DAG model so experts can provide estimates. The tool analyzes disagreement and aggregates conditional-probability assessments using both descriptive dispersion analysis and prior-regularized methods. Causal analysis can then provide insights into the DAG model, for example, via d-separation, adjustment-set inspection, do-calculus for what-if analysis, local independence checks, evidence updating, and impact-oriented searches for effective interventions. This workflow is illustrated with an instant-payments gateway scenario, demonstrating (a) explicit safe-state semantics, (b) Bowtie-to-DAG operationalization, (c) probability capture with visible expert noise, and (d) causal what-if analysis on a transformed and enriched model. Rather than presenting a statistical validation, the paper contributes a method and prototype system that transforms partially mitigated risks into observable, probabilistic, and intervention-ready models.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The paper presents the Hagenberg Risk Management Process as a method to operationalize qualitative Bowtie risk diagrams into formal probabilistic runtime models. It introduces Realtime Risk Studio, which transforms Bowtie elements (causes, top event, barriers, consequences) into DAGs for Bayesian inference while adding safe-state semantics and Activation Nodes as intervention points. A complementary Probability Capture tool generates expert questionnaires from the DAG, analyzes disagreement via dispersion and prior-regularized aggregation, and supports causal analysis (d-separation, do-calculus, adjustment sets). The workflow is illustrated with an instant-payments gateway scenario demonstrating explicit safe states, Bowtie-to-DAG conversion, visible expert noise, and what-if interventions. The contribution is explicitly methodological and prototype-based, without statistical validation.

Significance. If the Bowtie-to-DAG mapping preserves barrier logic and independence structure, and if expert aggregation proves reliable, the approach could provide a practical bridge from static qualitative risk assessment to observable, intervention-ready probabilistic models in operational settings such as financial gateways. The explicit inclusion of safe-state semantics and Activation Nodes, together with integrated causal tools, strengthens the potential utility for real-time risk monitoring.

major comments (1)
  1. [Abstract and Realtime Risk Studio] Abstract and Realtime Risk Studio section: The central claim requires that Bowtie structures can be converted into a DAG whose CPTs and d-separation structure support valid Bayesian inference and do-calculus. No explicit mapping algorithm, parent-child rules for barriers, or CPT construction procedure is supplied; the instant-payments example demonstrates the result but does not show how AND/OR barrier combinations or multi-pathway logic are encoded. Without these rules it is impossible to confirm that the resulting network recovers the original risk-pathway independencies or that interventions remain semantically faithful.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback and for recognizing the methodological focus of the work. We address the single major comment below and will revise the manuscript to provide the requested formal details.

read point-by-point responses

  1. Referee: [Abstract and Realtime Risk Studio] Abstract and Realtime Risk Studio section: The central claim requires that Bowtie structures can be converted into a DAG whose CPTs and d-separation structure support valid Bayesian inference and do-calculus. No explicit mapping algorithm, parent-child rules for barriers, or CPT construction procedure is supplied; the instant-payments example demonstrates the result but does not show how AND/OR barrier combinations or multi-pathway logic are encoded. Without these rules it is impossible to confirm that the resulting network recovers the original risk-pathway independencies or that interventions remain semantically faithful.

    Authors: We agree that an explicit, step-by-step mapping algorithm is needed to allow readers to verify preservation of barrier logic, independence structure, and intervention semantics. The current Realtime Risk Studio section and the instant-payments example describe the conceptual transformation and resulting DAG but do not supply formal parent-child rules or a CPT-construction procedure for AND/OR combinations and multi-pathway logic. In the revised manuscript we will insert a new subsection that (i) states the parent-child assignment rules for each Bowtie element, (ii) provides pseudocode for encoding barrier logic (including how AND/OR gates become deterministic or noisy-OR CPTs), (iii) shows how d-separation in the resulting DAG corresponds to the original risk-pathway independencies, and (iv) confirms that do-calculus interventions on Activation Nodes remain faithful to the safe-state semantics. These additions will make the conversion reproducible and the causal claims verifiable. revision: yes

Circularity Check

0 steps flagged

No circularity: methodological extension of standard techniques

full rationale

The paper introduces a new workflow and prototype tools (Realtime Risk Studio, Probability Capture) for transforming qualitative Bowtie diagrams into DAGs, adding safe-state semantics and Activation Nodes, then applying standard Bayesian inference, d-separation, do-calculus, and expert aggregation methods. No equations, fitted parameters, or predictions are defined in terms of themselves. The central claims rest on the explicit description of the transformation process and its illustration in the instant-payments example rather than on self-referential derivations or load-bearing self-citations. Prior parts of the Hagenberg process are referenced only for context; the operationalization, probability capture, and causal analysis steps are presented as independent contributions built on external standards (Bowtie diagrams, Bayesian networks, Pearl's causal framework). This is a self-contained method paper with no reduction of results to inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 3 invented entities

The central claim rests on domain assumptions about model transformations and expert elicitation without independent empirical support or formal verification in the abstract.

axioms (2)
  • domain assumption Bowtie diagrams can be represented as directed acyclic graphs suitable for Bayesian inference
    This is the core transformation assumed in Realtime Risk Studio.
  • domain assumption Expert estimates of conditional probabilities can be reliably elicited via questionnaires and aggregated despite disagreement
    Assumed for the Probability Capture tool.
invented entities (3)
  • Activation Nodes no independent evidence
    purpose: Designated intervention points in the risk model
    New concept introduced to extend Bowtie models.
  • Realtime Risk Studio no independent evidence
    purpose: Domain-specific tool for Bowtie to DAG transformation
    New software tool proposed.
  • Probability Capture no independent evidence
    purpose: Tool for generating questionnaires and aggregating expert probabilities
    New tool for probability elicitation.

pith-pipeline@v0.9.0 · 5611 in / 1494 out tokens · 127964 ms · 2026-05-10T17:33:59.180364+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

51 extracted references · 51 canonical work pages

  1. [1]

    Hagenberg Risk Management Process (Part 1): Multidimensional Polar Heatmaps for Context-Sensitive Risk Analysis,

    E. Hermann and H. Lampesberger, “Hagenberg Risk Management Process (Part 1): Multidimensional Polar Heatmaps for Context-Sensitive Risk Analysis,”arXiv:2601.07644, 2026

    work page arXiv 2026

  2. [2]

    Hagenberg Risk Management Process (Part 2): From Context-Sensitive Triage to Case Analysis With Bowtie and Bayesian Networks,

    E. Hermann and H. Lampesberger, “Hagenberg Risk Management Process (Part 2): From Context-Sensitive Triage to Case Analysis With Bowtie and Bayesian Networks,”arXiv:2602.19270, 2026

    work page arXiv 2026

  3. [3]

    Regulation (EU) 2022/2554 of the European Parlia- ment and of the Council of 14 December 2022 on digital operational resilience for the financial sector (DORA),

    European Union, “Regulation (EU) 2022/2554 of the European Parlia- ment and of the Council of 14 December 2022 on digital operational resilience for the financial sector (DORA),”Official Journal L 333, 2022

    work page 2022

  4. [4]

    Information Security Continuous Monitoring (ISCM) for Federal Information Systems and Organizations,

    K. L. Dempsey, N. S. Chawla, L. A. Johnson, R. Johnston, A. C. Jones, A. D. Orebaugh, M. A. Scholl, and K. M. Stine, “Information Security Continuous Monitoring (ISCM) for Federal Information Systems and Organizations,” NIST Special Publication 800-137, 2011

    work page 2011

  5. [5]

    The bowtie method: A review,

    A. de Ruijter and F. Guldenmund, “The bowtie method: A review,” Safety Science, vol. 88, pp. 211–218, 2016

    work page 2016

  6. [6]

    IEC 31010:2019 Risk management—Risk assessment techniques,

    International Electrotechnical Commission, “IEC 31010:2019 Risk management—Risk assessment techniques,” IEC Standard, 2019

    work page 2019

  7. [7]

    Dynamic safety analysis of process systems by mapping bow-tie into Bayesian network,

    N. Khakzad, F. Khan, and P. Amyotte, “Dynamic safety analysis of process systems by mapping bow-tie into Bayesian network,”Process Safety and Environmental Protection, vol. 91, no. 1–2, pp. 46–53, 2013

    work page 2013

  8. [8]

    Dynamic risk analysis using bow-tie approach,

    N. Khakzad, F. Khan, and P. Amyotte, “Dynamic risk analysis using bow-tie approach,”Reliability Engineering & System Safety, vol. 104, pp. 36–44, 2012

    work page 2012

  9. [9]

    pyBNBowtie: Python library for Bow-Tie Analysis based on Bayesian Networks,

    F. T. Zurheide, E. Hermann, and H. Lampesberger, “pyBNBowtie: Python library for Bow-Tie Analysis based on Bayesian Networks,” Procedia Computer Science, vol. 180, pp. 344–351, 2021

    work page 2021

  10. [10]

    Koller and N

    D. Koller and N. Friedman,Probabilistic Graphical Models: Principles and Techniques. Cambridge, MA, USA: MIT Press, 2009

    work page 2009

  11. [11]

    Fenton and M

    N. Fenton and M. Neil,Risk Assessment and Decision Analysis with Bayesian Networks, 2nd ed. Boca Raton, FL, USA: CRC Press, 2018

    work page 2018

  12. [12]

    Pearl,Causality: Models, Reasoning, and Inference, 2nd ed

    J. Pearl,Causality: Models, Reasoning, and Inference, 2nd ed. Cam- bridge, U.K.: Cambridge Univ. Press, 2009

    work page 2009

  13. [13]

    Robust causal inference using directed acyclic graphs: the R package ‘dagitty’,

    J. Textor, B. van der Zander, M. S. Gilthorpe, M. Li ´skiewicz, and G. T. H. Ellison, “Robust causal inference using directed acyclic graphs: the R package ‘dagitty’,”International Journal of Epidemiology, vol. 45, no. 6, pp. 1887–1894, 2016

    work page 2016

  14. [14]

    pgmpy: A Python Toolkit for Bayesian Networks,

    A. Ankan and J. Textor, “pgmpy: A Python Toolkit for Bayesian Networks,”Journal of Machine Learning Research, vol. 25, no. 265, pp. 1–8, 2024

    work page 2024

  15. [15]

    Learning Bayesian Networks with the bnlearn R Package,

    M. Scutari, “Learning Bayesian Networks with the bnlearn R Package,” Journal of Statistical Software, vol. 35, no. 3, pp. 1–22, 2010

    work page 2010

  16. [16]

    SMILE: Structural Modeling, Inference, and Learn- ing Engine and GeNIe: A Development Environment for Graphical Decision-Theoretic Models,

    M. J. Druzdzel, “SMILE: Structural Modeling, Inference, and Learn- ing Engine and GeNIe: A Development Environment for Graphical Decision-Theoretic Models,” inProc. AAAI/IAAI, 1999, pp. 902–903

    work page 1999

  17. [17]

    The Hugin Tool for Learning Bayesian Networks,

    A. L. Madsen, M. Lang, U. B. Kjærulff, and F. Jensen, “The Hugin Tool for Learning Bayesian Networks,” inSymbolic and Quantitative Ap- proaches to Reasoning with Uncertainty, T. D. Nielsen and N. L. Zhang, Eds. Berlin, Germany: Springer, 2003, pp. 594–605

    work page 2003

  18. [18]

    ISO 31000:2018 Risk management—Guidelines,

    International Organization for Standardization, “ISO 31000:2018 Risk management—Guidelines,” ISO Standard, 2018

    work page 2018

  19. [19]

    A Bayesian approach to construct bow tie diagrams for risk evaluation,

    A. Badreddine and N. Ben Amor, “A Bayesian approach to construct bow tie diagrams for risk evaluation,”Process Safety and Environmental Protection, vol. 91, no. 3, pp. 159–171, 2013

    work page 2013

  20. [20]

    Make Bow-tie Dynamic by Rethinking It as a Hierarchical Bayesian Network: Dynamic Risk Assessment of an LNG Bunkering Operation,

    T. Vairo, A. C. Benvenuto, A. Tedeschi, A. P. Reverberi, and B. Fabiano, “Make Bow-tie Dynamic by Rethinking It as a Hierarchical Bayesian Network: Dynamic Risk Assessment of an LNG Bunkering Operation,” Chemical Engineering Transactions, vol. 91, pp. 277–282, 2022

    work page 2022

  21. [21]

    A situation risk awareness approach for process systems safety,

    M. Naderpour, J. Lu, and G. Zhang, “A situation risk awareness approach for process systems safety,”Safety Science, vol. 64, pp. 173–189, 2014

    work page 2014

  22. [22]

    A framework for dynamic risk assess- ment with condition monitoring data and inspection data,

    J. Xing, Z. Zeng, and E. Zio, “A framework for dynamic risk assess- ment with condition monitoring data and inspection data,”Reliability Engineering & System Safety, vol. 191, p. 106552, 2019

    work page 2019

  23. [23]

    Integration of deep learning and Bayesian networks for condition and operation risk monitoring of complex engineering systems,

    R. Moradi, S. Cofre-Martel, E. Lopez Droguett, M. Modarres, and K. M. Groth, “Integration of deep learning and Bayesian networks for condition and operation risk monitoring of complex engineering systems,”Reliability Engineering & System Safety, vol. 222, p. 108433, 2022

    work page 2022

  24. [24]

    Toward a framework for risk monitoring of complex engineering systems with online operational data: A deep learning-based solution,

    R. Moradi, A. Ruiz-Tagle Palazuelos, E. Lopez Droguett, and K. M. Groth, “Toward a framework for risk monitoring of complex engineering systems with online operational data: A deep learning-based solution,”Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability, vol. 237, no. 5, pp. 910–921, 2023

    work page 2023

  25. [25]

    The future of risk assessment,

    E. Zio, “The future of risk assessment,”Reliability Engineering & System Safety, vol. 177, pp. 176–190, 2018

    work page 2018

  26. [26]

    Overview on Bayesian networks applications for dependability, risk analysis and maintenance areas,

    P. Weber, G. Medina-Oliva, C. Simon, and B. Iung, “Overview on Bayesian networks applications for dependability, risk analysis and maintenance areas,”Engineering Applications of Artificial Intelligence, vol. 25, no. 4, pp. 671–682, 2012

    work page 2012

  27. [27]

    Bayesian networks in reliability,

    H. Langseth and L. Portinale, “Bayesian networks in reliability,”Relia- bility Engineering & System Safety, vol. 92, no. 1, pp. 92–108, 2007

    work page 2007

  28. [28]

    Applications of Bayesian networks and Petri nets in safety, reliability, and risk assessments: A review,

    S. Kabir and Y . Papadopoulos, “Applications of Bayesian networks and Petri nets in safety, reliability, and risk assessments: A review,”Safety Science, vol. 115, pp. 154–175, 2019

    work page 2019

  29. [29]

    Evolution of Safety and Security Risk Assessment methodologies towards the use of Bayesian Networks in Process Industries,

    P. G. George and V . R. Renjith, “Evolution of Safety and Security Risk Assessment methodologies towards the use of Bayesian Networks in Process Industries,”Process Safety and Environmental Protection, vol. 149, pp. 758–775, 2021

    work page 2021

  30. [30]

    Application of Bayesian Networks in Quantitative Risk Assessment of Subsea Blowout Preventer Operations,

    B. Cai, Y . Liu, Z. Liu, X. Tian, Y . Zhang, and R. Ji, “Application of Bayesian Networks in Quantitative Risk Assessment of Subsea Blowout Preventer Operations,”Risk Analysis, vol. 33, no. 7, pp. 1293–1311, 2013

    work page 2013

  31. [31]

    A Bayesian Networks approach to Operational Risk,

    V . Aquaro, M. Bardoscia, R. Bellotti, A. Consiglio, F. De Carlo, and G. Ferri, “A Bayesian Networks approach to Operational Risk,”Physica A: Statistical Mechanics and its Applications, vol. 389, no. 8, pp. 1721– 1728, 2010

    work page 2010

  32. [32]

    Modernising operational risk management in financial institutions via data-driven causal factors analysis: A pre-registered study,

    N. Cornwell, C. Bilson, A. Gepp, S. Stern, and B. J. Vanstone, “Modernising operational risk management in financial institutions via data-driven causal factors analysis: A pre-registered study,”Pacific-Basin Finance Journal, vol. 79, p. 102011, 2023

    work page 2023

  33. [33]

    O’Hagan, C

    A. O’Hagan, C. E. Buck, A. Daneshkhah, J. R. Eiser, P. H. Garthwaite, D. J. Jenkinson, J. E. Oakley, and T. Rakow,Uncertain Judgements: Eliciting Experts’ Probabilities. Chichester, U.K.: Wiley, 2006

    work page 2006

  34. [34]

    Expert judgement for dependence in probabilistic modelling: A systematic literature review and future research directions,

    C. Werner, T. Bedford, R. M. Cooke, A. M. Hanea, and O. Morales- Nápoles, “Expert judgement for dependence in probabilistic modelling: A systematic literature review and future research directions,”European Journal of Operational Research, vol. 258, no. 3, pp. 801–819, 2017

    work page 2017

  35. [35]

    A proposed validation framework for expert elicited Bayesian Networks,

    J. Pitchforth and K. Mengersen, “A proposed validation framework for expert elicited Bayesian Networks,”Expert Systems with Applications, vol. 40, no. 1, pp. 162–167, 2013

    work page 2013

  36. [36]

    Sensitivity and ro- bustness analysis in Bayesian networks with the bnmonitor R package,

    M. Leonelli, R. Ramanathan, and R. L. Wilkerson, “Sensitivity and ro- bustness analysis in Bayesian networks with the bnmonitor R package,” Knowledge-Based Systems, vol. 278, p. 110882, 2023

    work page 2023

  37. [37]

    Pearl,Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference

    J. Pearl,Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. San Mateo, CA, USA: Morgan Kaufmann, 1988

    work page 1988

  38. [38]

    Darwiche,Modeling and Reasoning with Bayesian Networks

    A. Darwiche,Modeling and Reasoning with Bayesian Networks. Cam- bridge, U.K.: Cambridge Univ. Press, 2009

    work page 2009

  39. [39]

    Local computations with probabilities on graphical structures and their application to expert systems,

    S. L. Lauritzen and D. J. Spiegelhalter, “Local computations with probabilities on graphical structures and their application to expert systems,”Journal of the Royal Statistical Society: Series B, vol. 50, no. 2, pp. 157–224, 1988

    work page 1988

  40. [40]

    Spirtes, C

    P. Spirtes, C. Glymour, and R. Scheines,Causation, Prediction, and Search, 2nd ed. Cambridge, MA, USA: MIT Press, 2000

    work page 2000

  41. [41]

    Causal discovery and inference: concepts and recent methodological advances,

    P. Spirtes and K. Zhang, “Causal discovery and inference: concepts and recent methodological advances,”Applied Informatics, vol. 3, Art. no. 3, 2016

    work page 2016

  42. [42]

    Review of causal discovery methods based on graphical models,

    C. Glymour, K. Zhang, and P. Spirtes, “Review of causal discovery methods based on graphical models,”Frontiers in Genetics, vol. 10, p. 524, 2019

    work page 2019

  43. [43]

    Causal Inference Using Graphical Models with the R Package pcalg,

    M. Kalisch, M. Mächler, D. Colombo, M. H. Maathuis, and P. Bühlmann, “Causal Inference Using Graphical Models with the R Package pcalg,”Journal of Statistical Software, vol. 47, no. 11, pp. 1– 26, 2012

    work page 2012

  44. [44]

    Identifying Causal Effects with the R Package causaleffect,

    S. Tikka and J. Karvanen, “Identifying Causal Effects with the R Package causaleffect,”Journal of Statistical Software, vol. 76, no. 12, pp. 1–30, 2017

    work page 2017

  45. [45]

    Causal-learn: Causal Discovery in Python,

    Y . Zheng, B. Huang, W. Chen, J. Ramsey, M. Gong, R. Cai, S. Shimizu, P. Spirtes, and K. Zhang, “Causal-learn: Causal Discovery in Python,” Journal of Machine Learning Research, vol. 25, no. 60, pp. 1–7, 2024

    work page 2024

  46. [46]

    Causal Discovery Toolbox: Uncovering causal relationships in Python,

    D. Kalainathan, O. Goudet, and R. Dutta, “Causal Discovery Toolbox: Uncovering causal relationships in Python,”Journal of Machine Learn- ing Research, vol. 21, no. 37, pp. 1–5, 2020

    work page 2020

  47. [47]

    Python package for causal discovery based on LiNGAM,

    T. Ikeuchi, M. Ide, Y . Zeng, T. N. Maeda, and S. Shimizu, “Python package for causal discovery based on LiNGAM,”Journal of Machine Learning Research, vol. 24, no. 14, pp. 1–8, 2023

    work page 2023

  48. [48]

    Huntington-Klein,The Effect: An Introduction to Research Design and Causality

    N. Huntington-Klein,The Effect: An Introduction to Research Design and Causality. Boca Raton, FL, USA: CRC Press, 2021

    work page 2021

  49. [49]

    Kahneman, O

    D. Kahneman, O. Sibony, and C. R. Sunstein,Noise: A Flaw in Human Judgment. New York, NY , USA: Little, Brown Spark, 2021

    work page 2021

  50. [50]

    Black Swans and the Domains of Statistics,

    N. N. Taleb, “Black Swans and the Domains of Statistics,”The American Statistician, vol. 61, no. 3, pp. 198–200, 2007

    work page 2007

  51. [51]

    N. N. Taleb,The Black Swan: The Impact of the Highly Improbable, 2nd ed. New York, NY , USA: Random House Trade Paperbacks, 2010

    work page 2010