FIELD NOTES / Clinical Data Science
DAG-Dowhy
CompletedEvidence B
Auditable binary treatment/outcome estimation from an explicit DAG.
Last reviewed 2026-10-19 · Public research / code
Overview
Causal inference · DAG-Dowhy
Auditable binary treatment/outcome estimation from an explicit DAG.
Research Question
Auditable binary treatment/outcome estimation from an explicit DAG.
Technical Approach
Cross-fitted AIPW risk differences, cycle/backdoor checks, propensity and balance diagnostics.
PythonNumPypandasSciPy
My Contribution
Developed the public pipeline and synthetic tests.
Evidence
- Repository core, dependency and test/development records inspected during the 2026-10-19 audit. Evidence levels describe the available record, not clinical effectiveness.
Outputs
- Code and documented development artifacts. Private source and study data are not redistributed.
Current Status
CompletedEvidence B
Development evidence is available; no claim of clinical utility or production readiness is made.
Related Publications
No project-specific peer-reviewed publication is listed. Ongoing work is kept separate from the published bibliography.