Simulation infrastructure for constructing, simulating, and evaluating causal data-generating processes with known ground truth. Covariate roles are first-class concepts in the API, and every evaluation returns bias, RMSE, coverage, and power with Monte Carlo standard errors. Designed for benchmarking causal estimators and building reproducible simulation studies.
Weighted Multi-Study Analysis Package for causal inference across multiple observational studies. Implements three weighting approaches including FLEXOR, which maximizes effective sample size through iterative optimization, to enable valid estimation of group-specific potential outcomes.
A novel statistical framework for borrowing information from external cohorts to augment inference in a target study. Identifies importance weights by maximizing effective sample size subject to covariate balance constraints, explicitly controlling for distributional shift between populations.
Generates randomized practice questions for learning statistics concepts and methods. Covers hypothesis testing, regression, causal inference, and probability with auto-graded interactive prompts for self-study and teaching.
Lord of the Rings themed color palettes for ggplot2. Provides a collection of palettes drawn from the visual identity of Middle-earth, suitable for both categorical and continuous data visualization.
Generates reproducible research project scaffolds with standardized folder structures, configured Jupyter notebooks, and auto-populated READMEs. Reduces setup overhead and enforces consistent organization across projects.
Retrieves papers from academic APIs, extracts key metadata, and generates concise literature scan reports. Designed for rapid scoping reviews and systematic evidence gathering.
A code-first reference for applied causal inference and causal ML methods. Each estimator includes identification assumptions, estimation strategy, robustness considerations, and tabbed R and Python implementation code. Includes an estimator decision tree.
Full-stack platform that maps health services and policy research as an interactive knowledge graph, tagging each paper's identification strategy, policy lever, and outcomes using natural language processing and curated method taxonomies. Supports exploration of a literature by causal design (e.g., difference-in-differences, regression discontinuity), with sentence-level evidence for every tag, filtering by method, policy lever, and outcome, and export of filtered results to CSV and PDF.
Interactive browser tool for constructing and reasoning about directed acyclic graphs as graphical causal models. Supports visual DAG building, automatic identification of adjustment sets, and backdoor path analysis. Runs entirely client-side.