Bounding Treatment Effects by Pooling Limited Information across Observations
-
Updated
May 11, 2026 - R
Bounding Treatment Effects by Pooling Limited Information across Observations
Causal Inference in Case-Control Studies
Cross-asset price impact is set-identified, not point-identified: the confounding gap has rank at most K + rank(B). A diagonal truth still yields a dense cross-impact matrix, and the trades immune to the resulting cost error form a large but measure-zero subspace. Preregistered and reproducible.
Reproducibility artifact for 'Sharp Target-Domain Certificates for Quantum-Kernel Advantage under Distribution Shift': sharp finite-batch certificates, controlled quantum-vs-classical kernel benchmark (EMBER, UNSW-NB15, ToN-IoT, TableShift), frozen specifications, prediction locks, manuscript sources. Zenodo v1.1.6.
[Experimental] Federated Partial Identification for Causal Inference with OMOP CDM
Stata package for difference-in-discontinuities (Picchetti–Pinto–Shinoki 2026): local polynomial estimation when a confounder sits at the same cutoff, plus two validity tests and partial-identification bounds.
Claim-relative identified sets across benchmark evaluation specifications
Abstention and bias scores in BBQ-style benchmarks: code, data and results
A unified Python framework for causal effect bounding algorithms
Python library that bounds the failure rate of stacked AI guardrails from each rail's individual rate, instead of assuming their failures are independent.
Implementation for the working paper: Symmetry-Informed Causal Partial Identification
Estimating the Effect of Persuasion in Stata
Archival code, results, and reproducibility bundle for common-witness counterfactual image audits; see KNOWN_ISSUES.md before rerunning.
Measuring the cost of zero labelling probability in global crop mapping: why prediction-powered inference fails across countries, and how overlap-gated bounds, national statistics, and minimal target sampling restore validity.
R package for quantile-only regression: inference and partial identification from disclosed group quantiles.
Pre-registered positivity audit of global carbon-flux upscaling (MetaFlux). Measures where gridded land-months fall outside eddy-covariance tower support, tests whether ensemble uncertainty notices extrapolation, and derives sharp Lipschitz bounds on global GPP.
Mémoire d'actuariat (ENSAE Paris, Institut des actuaires) : le besoin de capital lié à la non-conformité au règlement DORA, modélisé par une cascade dirigée entre ses cinq piliers. Chaque nombre du mémoire est vérifié contre la sortie du script qui le calcule.
Reproducible code and derived data for testing how present-day stellar constraints identify cumulative EUV and wind histories in TOI-700 and LHS 1140.
What 12 million powerlifting attempts can and cannot say about attempt selection. Paper, code and a free attempt calculator.
Estimating the Effect of Persuasion in Stata
To associate your repository with the partial-identification topic, visit your repo's landing page and select "manage topics."