REVIEW 17 cited by
DoWhy: An End-to-End Library for Causal Inference
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
In addition to efficient statistical estimators of a treatment's effect, successful application of causal inference requires specifying assumptions about the mechanisms underlying observed data and testing whether they are valid, and to what extent. However, most libraries for causal inference focus only on the task of providing powerful statistical estimators. We describe DoWhy, an open-source Python library that is built with causal assumptions as its first-class citizens, based on the formal framework of causal graphs to specify and test causal assumptions. DoWhy presents an API for the four steps common to any causal analysis---1) modeling the data using a causal graph and structural assumptions, 2) identifying whether the desired effect is estimable under the causal model, 3) estimating the effect using statistical estimators, and finally 4) refuting the obtained estimate through robustness checks and sensitivity analyses. In particular, DoWhy implements a number of robustness checks including placebo tests, bootstrap tests, and tests for unoberved confounding. DoWhy is an extensible library that supports interoperability with other implementations, such as EconML and CausalML for the the estimation step. The library is available at https://github.com/microsoft/dowhy
Forward citations
Cited by 17 Pith papers
-
A cubical formalisation of topos causal models: intervention, forcing, and a contextuality obstruction
A machine-checked Cubical Agda formalisation of topos causal models proves the intervention classifier, sheaf gluing, forcing clauses, and a corrected Lawvere-Tierney do-calculus, and adds a verified contextuality obs...
-
A cubical formalisation of conditional independence, Bayesian conditioning, and Pearl's d-separation soundness
In Cubical Agda, conditional independence is formalized as paths between kernels, Bayesian conditioning is made total on a full-support fragment, and Pearl's d-separation soundness is verified for finite DAGs, using a...
-
ECLAIR: A Causally-Grounded AI Framework for Scientific Discovery in Empirical Software Engineering
ECLAIR integrates LLMs into a human-in-the-loop, causal-inference-based scientific workflow for software engineering, and a pilot study finds small negative causal effects of certain prompt designs on code generation ...
-
Reason-Mediated Behavioral Models for Auditing LLM Social Simulators
Signed rationale-derived reason states improve held-out purchase-intent prediction in a 94-person concept test, while LLM-simulated reasons fail the same human readout.
-
Root Cause Analysis of Outliers in Unknown Cyclic Graphs
Applying the inverse covariance (precision) matrix of normal data to one anomalous sample exposes the root cause plus its cycle-involved parents in linear cyclic systems, without knowing the graph.
-
Causal identification with $Y_0$
Y0 is an open-source Python package implementing a broad suite of causal identification algorithms with a domain-specific language for queries and estimands.
-
Measurement as Bricolage: Examining How Data Scientists Construct Target Variables for Predictive Modeling Tasks
Data scientists construct prediction targets through bricolage, applying five reformulation strategies (piggybacking, composing, swapping, bridging, refining) to balance five criteria: validity, simplicity, predictabi...
-
Causality-aware Safety Testing for Autonomous Driving Systems
Causal-Fuzzer uses causal graphs of scene, action, and violation relationships to guide simulation fuzzing, and reports finding more diverse violations and better testing sufficiency than three baselines on Apollo.
-
FairPFN: A Tabular Foundation Model for Causal Fairness
FairPFN pre-trains a transformer on synthetic structural-causal-model data so it can remove the causal effect of a binary protected attribute from tabular predictions without a user-supplied causal graph.
-
Understanding the European energy crisis through structural causal models
A causal analysis of French electricity prices shows the 2021-2023 price surge was driven mainly by rising gas prices through cross-border coupling and, secondarily, by nuclear outages, and quantifies the indirect eff...
-
proxymate: Diagnosis and Adjustment of Proxy Estimates for Reliable Inference
A four-level diagnostic-to-adjustment framework (representativity, unit, estimate, domain) plus open-source package maps proxy failures to corrections like IW, PPI, calibration, and meta-analytic recalibration, with M...
-
A Causal Perspective on Measuring, Explaining and Mitigating Smells in LLM-Generated Code
A probabilistic score of code-smell propensity in LLM output is validated, used in a causal analysis, and shown to drop when prompts explicitly discourage known smells.
-
What Causes COVID-19 Fear? General Drivers of Fear During a Health Crisis
The paper claims information source choice dominates age and education as a correlate of COVID-19 fear in US survey data, and that source reliance clusters US states by political lean.
-
From Observational Data to Clinical Recommendations: A Causal Framework for Estimating Patient-level Treatment Effects and Learning Policies
A causal framework for learning treatment policies with deferral, applied to diuretic dosing in acute heart failure with kidney injury.
-
Causal-TS: A Python Library for Causal Discovery in High-Dimensional and Nonstationary Time Series
Causal-TS ships CDNOTS, CDNOTS+, CEDAR, and GRACE plus baselines under one GPU-accelerated CI-test API with regime discovery and DoWhy hooks for nonstationary time-series causal graphs.
-
teLLMe Why (Ain't Nothing but a Jam): Exploratory Causal Analysis of Urban Driving Data
A causal-analysis toolbox for dashcam-derived traffic events: PC-learned graphs, LLM-parsed questions, and regression/DoWhy effect estimates bundled into explainable "Causal Cards."
-
From Images to Insights: Explainable Biodiversity Monitoring with Plain Language Habitat Explanations
An image-to-text pipeline that combines species recognition, occurrence data, and causal inference to produce plain-language habitat preference explanations, demonstrated on one bee and one flower species.
Discussion (0). Sign in to comment.