Software

Stata

A Stata package implementing the conditional extrapolation pre-test framework for difference-in-differences designs proposed by Mikhaeil and Harshaw (2025). Provides an asymptotically consistent pre-test for the extrapolation condition and conditionally valid confidence intervals for the ATT with guaranteed asymptotic coverage.

Stata

A Stata implementation of equivalence tests for pre-trends in DiD estimation, based on Dette and Schumann (2024, JBES). Implements three equivalence hypotheses (maximum, mean, and RMS), minimum equivalence threshold computation, multiple inference methods including spherical and wild bootstrap, and visualization with equivalence bounds.

Stata

A Stata implementation of the Double Difference-in-Differences method proposed by Egami and Yamauchi (2023, Political Analysis). Optimally combines standard DID and sequential DID via GMM for improved efficiency and robustness. Supports staggered adoption designs, parallel trends diagnostics with equivalence confidence intervals, and visualization.

Stata · Rust backend

Stata implementation of the Triply Robust Panel (TROP) estimator from Athey, Imbens, Qu & Viviano (2025). Combines unit weights, time weights, and nuclear-norm-regularized low-rank regression adjustment to estimate ATT in panel data. Supports twostep and joint estimation, LOOCV tuning, bootstrap inference, and general assignment patterns. High-performance Rust backend.

Stata

Stata package for doubly robust uniform confidence bands for group-time conditional average treatment effects (CATT) in staggered DiD. Implements Imai, Qin, and Yanagi (2025, JBES). Combines IPW with outcome regression via local polynomial smoothing, with flexible aggregation into event-study, group, calendar, and simple summary parameters.

Stata

Doubly robust semiparametric DiD estimator for high-dimensional covariates, based on Ning, Peng, and Tao (2020). Features cross-fitted AIPW estimation, Lasso penalization, CLIME debiasing, polynomial/trigonometric sieve bases, and bootstrap uniform confidence bands.

Stata

Stata implementation of Weighted Fixed Effects (WFE) and Propensity-score Weighted Fixed Effects (PWFE) estimators from Imai & Kim (2021, Political Analysis). Derives observation-specific weights targeting ATE/ATT, supports one-way/two-way FE, first-difference, matched DiD, and includes White misspecification test.

R

R package implementing the Lee & Wooldridge (2025, 2026) rolling DiD estimator for panel data. Supports common timing and staggered adoption designs, four estimators (RA/IPW/IPWRA/PSM), Fisher randomization inference, wild cluster bootstrap, rich diagnostics, and LaTeX/CSV export. Built on data.table for high performance.

Python · PyPI

A Python implementation of the Lee–Wooldridge (2025, 2026) rolling-transformation approach to difference-in-differences estimation. Supports common timing and staggered adoption designs, exact small-sample inference, seasonal adjustments, IPW/IPWRA/PSM estimators, pre-treatment dynamics testing, and comprehensive diagnostic toolkits.

Python · PyPI

Python implementation of Covariate Balancing Propensity Score (CBPS) for robust causal inference. Supports binary, multi-valued, and continuous treatments via GMM framework. Includes hdCBPS, npCBPS, marginal structural models (CBMSM), instrumental variables (CBIV), and comprehensive diagnostics. Numerically accurate within ±1e-6 of the CBPS R package.

OpenAI Codex · Multi-Agent Framework

A multi-agent workflow framework for OpenAI Codex that builds, tests, and documents statistical software packages with adversarial verification. Implements the StatsClaw methodology (Qin & Xu, 2026) with 8 specialized AI roles operating under strict information barriers. Adapted from the Claude Code edition.

AI Skill · Academic Writing

A systematic AI skill for writing R Journal econometrics papers — covering the full workflow from draft to submission. Provides structured guidance on paper architecture, code examples, reproducibility, and R Journal formatting standards.