Does it handle non-centered parameterization?

Absolutely. It provides specific code patterns for non-centered parameterization to help resolve divergent transitions often found in hierarchical models.

Can this skill help with MCMC diagnostics?

Yes, it includes specific guidance on monitoring Rhat, Effective Sample Size (ESS), divergences, and performing posterior predictive checks.

What version of Stan does this skill support?

It is optimized for Stan 2.37 and newer, utilizing modern syntax such as the 'array' keyword and updated distribution parameterizations.

Is this skill compatible with R integration?

Yes, it includes templates for cmdstanr, allowing you to sample, diagnose, and compare models (LOO/WAIC) directly within an R environment.

Stan Bayesian Modeling Fundamentals

Name: Stan Bayesian Modeling Fundamentals
Author: choxos

bychoxos

0•

Data Science & ML

Provides foundational knowledge and best practices for developing, optimizing, and reviewing Stan 2.37 probabilistic models.

This skill equips Claude with the specific syntax, structural requirements, and statistical best practices needed to build robust Bayesian models in Stan. It covers the essential seven-block program structure, the modern array syntax, constrained variable types, and efficient vectorization techniques. Whether you are translating models from BUGS/JAGS or building new hierarchical models from scratch, this skill ensures the resulting code follows modern Stan conventions (version 2.37+) and integrates seamlessly with R-based workflows like cmdstanr and the bayesplot ecosystem for comprehensive model diagnostics.

Key Features

010 GitHub stars

02Library of continuous, discrete, and multivariate distributions

03Implementation patterns for vectorization and non-centered parameterization

04Integrated Bayesian workflow diagnostics including Rhat, ESS, and LOO

05Complete Stan 2.37 program block structure guidance

06Reference for constrained types and modern array syntax

Use Cases

01Writing new Bayesian hierarchical models from scratch

02Optimizing slow Stan code through vectorization and parameterization tricks

03Translating legacy BUGS/JAGS models to modern Stan syntax

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add choxos/biostatagent stan-fundamentals

For use in Claude.ai and ChatGPT

Key Features

010 GitHub stars

02Library of continuous, discrete, and multivariate distributions

03Implementation patterns for vectorization and non-centered parameterization

04Integrated Bayesian workflow diagnostics including Rhat, ESS, and LOO

05Complete Stan 2.37 program block structure guidance

06Reference for constrained types and modern array syntax

Use Cases

01Writing new Bayesian hierarchical models from scratch

02Optimizing slow Stan code through vectorization and parameterization tricks

03Translating legacy BUGS/JAGS models to modern Stan syntax

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add choxos/biostatagent stan-fundamentals

For use in Claude.ai and ChatGPT