Can I use this skill to compare different statistical models?

Yes, it provides standardized code for comparing models using information criteria like Leave-One-Out (LOO) cross-validation and WAIC, as well as model averaging techniques.

Is it compatible with the latest version of PyMC?

Yes, the skill is optimized for PyMC version 5.x and later, utilizing modern APIs and the ArviZ library for posterior analysis.

How does this skill handle model sampling issues?

The skill includes specific troubleshooting steps for common issues like divergences and low effective sample size (ESS), recommending techniques like non-centered parameterization and adjusting target acceptance rates.

What is the PyMC Bayesian Modeling skill?

It is a specialized capability for Claude that provides domain-specific guidance and code patterns for building, fitting, and diagnosing Bayesian models using the PyMC library.

Does this skill support hierarchical data structures?

Absolutely. It includes templates and best practices specifically for hierarchical/multilevel models, ensuring correct parameterization to avoid sampling difficulties.

PyMC Bayesian Modeling

Name: PyMC Bayesian Modeling
Author: plurigrid

byplurigrid

•

データサイエンスとML

Builds, fits, and validates sophisticated Bayesian statistical models using PyMC's probabilistic programming framework.

The PyMC Bayesian Modeling skill empowers Claude to execute comprehensive Bayesian workflows, ranging from data preparation and prior predictive checks to MCMC sampling using NUTS and variational inference. It provides structured implementation patterns for hierarchical modeling, regression, and time-series analysis, ensuring robust uncertainty quantification. By incorporating best practices for model diagnostics (like R-hat and ESS) and comparison metrics (LOO and WAIC), this skill helps data scientists and researchers build reliable probabilistic models that handle missing data and complex multilevel structures with statistical rigor.

主な機能

01Sophisticated hierarchical and multilevel model construction patterns

02Model comparison and averaging using LOO and WAIC criteria

03End-to-end Bayesian workflow automation from priors to posterior checks

04Support for modern PyMC 5.x API and ArviZ integration

05Comprehensive model diagnostics including R-hat, ESS, and divergence checks

068 GitHub stars

ユースケース

01Implementing hierarchical linear or logistic regression for grouped datasets

02Performing principled Bayesian inference on datasets with missing values or measurement errors

03Conducting uncertainty quantification for business forecasting and time-series

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add plurigrid/asi pymc

For use in Claude.ai and ChatGPT

Download Skill