How does it handle model comparison?

The skill guides users through using ArviZ for model comparison using Leave-One-Out (LOO) cross-validation and Widely Applicable Information Criterion (WAIC).

Can this skill help resolve MCMC sampling divergences?

Yes, it provides specific strategies for troubleshooting divergences, including adjusting target_accept parameters and implementing non-centered parameterizations.

Does it include support for hierarchical models?

Absolutely. It includes patterns for multilevel and hierarchical data structures, focusing on parameterization techniques that ensure sampling efficiency.

What version of PyMC does this skill support?

This skill is designed for modern PyMC (version 5.x+) APIs, ensuring compatibility with the latest probabilistic programming features and ArviZ integration.

PyMC Bayesian Modeling

Name: PyMC Bayesian Modeling
Author: Zehong-Wang

byZehong-Wang

0•

Data Science & ML

Build, fit, and validate complex Bayesian probabilistic models using the PyMC Python library and modern MCMC sampling techniques.

The PyMC Bayesian Modeling skill provides Claude with expert-level guidance for probabilistic programming and statistical inference. It facilitates the entire Bayesian workflow, including data preparation, prior selection, and model fitting using the No-U-Turn Sampler (NUTS) or Variational Inference (ADVI). The skill includes specialized patterns for hierarchical models, linear/logistic regression, and time series analysis, while ensuring rigorous validation through prior/posterior predictive checks and diagnostic metrics like R-hat and Effective Sample Size (ESS).

Key Features

01Automated model diagnostics including R-hat, ESS, and divergence troubleshooting

02Advanced MCMC sampling implementation using NUTS and variational inference via ADVI

03Model comparison and selection using LOO, WAIC, and Pareto-k values

04Hierarchical and multilevel model templates with non-centered parameterization

05Comprehensive Bayesian workflow management from data prep to posterior checks

060 GitHub stars

Use Cases

01Handling missing data and measurement errors through principled probabilistic modeling

02Quantifying uncertainty in scientific research and complex business forecasting

03Analyzing nested or grouped datasets using robust hierarchical Bayesian frameworks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add zehong-wang/kosmos pymc

For use in Claude.ai and ChatGPT

Download Skill