Code

Software Packages

horseshoe

Bayesian linear regression with the horseshoe prior (Carvalho, Polson, and Scott, 2010). Implements the horseshoe shrinkage estimator for high-dimensional regression problems where many coefficients are expected to be near zero.

Languages: Stata, R (R package name: hsreg)

GitHub: https://github.com/amceachin-code/horseshoe

Reference: Carvalho, C.M., Polson, N.G., & Scott, J.G. (2010). The horseshoe estimator for sparse signals. Biometrika, 97(2), 465-480.

rdtp

Searches over candidate cutoffs on a discrete forcing variable to find the cutoff maximizing R-squared from a linear-spline regression discontinuity design. Based on McEachin, Domina, and Penner (2020, JPAM).

Languages: Stata, R

GitHub: https://github.com/amceachin-code/rdtp

Reference: McEachin, A., Domina, T., & Penner, A. (2020). Heterogeneous effects of early algebra across California middle schools. Journal of Policy Analysis and Management, 39(3), 772-800.

Claude Code Skills

writing-style

A sentence-level prose editor grounded in Verlyn Klinkenborg’s Several Short Sentences About Writing. Revises text from inline input, a file, or a URL in three passes: clarity, agency, and structure.

Usage: /writing-style

GitHub: https://github.com/amceachin-code/claude_skills/tree/main/writing-style

datascience-reviewer

Code review from the perspective of a senior data scientist. Evaluates structure, correctness risks, efficiency, usability, and maintainability, tags each finding by severity, and ends with a prioritized list of concrete changes.

Usage: /datascience-reviewer

GitHub: https://github.com/amceachin-code/claude_skills/tree/main/datascience-reviewer

methods-reviewer

Methodological review from a causal inference and semiparametric statistics perspective. Challenges identification assumptions, estimation strategies, and finite-sample claims before work goes to expert audiences.

Usage: /methods-reviewer

GitHub: https://github.com/amceachin-code/claude_skills/tree/main/methods-reviewer