Does it provide guidance on choosing priors?

Yes, it includes a reference table for weakly informative priors based on Gelman et al. recommendations for intercepts, coefficients, and scale parameters.

What probabilistic programming languages are supported?

This skill provides implementations for both Stan and JAGS, the two most widely used languages for Bayesian inference and statistical modeling.

How do I implement count data models?

The skill provides templates for both Poisson regression (standard counts) and Negative Binomial regression, which is ideal for count data with overdispersion.

What is robust regression in this context?

Robust regression uses a Student-t distribution for the likelihood instead of a Normal distribution, allowing the model to be less sensitive to outliers by utilizing the 'nu' (degrees of freedom) parameter.

Can this skill help with correlated predictors?

Yes, it includes specialized QR decomposition templates in Stan to improve sampling efficiency and numerical stability when dealing with highly correlated independent variables.

Bayesian Regression Models

Name: Bayesian Regression Models
Author: choxos

bychoxos

0•

Data Science & ML

Implements advanced Bayesian regression models including linear, logistic, and robust variations using Stan and JAGS.

This skill provides specialized guidance and implementation patterns for Bayesian modeling, offering production-ready templates for Stan and JAGS. It covers a wide range of regression types such as linear, logistic, Poisson, and negative binomial, along with robust regression techniques using Student-t errors and QR decomposition for handling correlated predictors. By providing standardized code blocks and expert recommendations for weakly informative priors based on established statistical research, it enables researchers and data scientists to build reliable statistical models directly within their Claude-assisted workflow.

Key Features

01Performance optimization using QR decomposition for correlated predictors

02Expert recommendations for weakly informative prior distributions

03Specialized models for overdispersed counts and robust Student-t regression

040 GitHub stars

05Comprehensive templates for Linear, Logistic, and Poisson regression

06Support for Stan and JAGS probabilistic programming languages

Use Cases

01Building hierarchical or complex statistical models for biostatistics and academic research

02Converting frequentist models into a Bayesian framework for superior uncertainty quantification

03Implementing robust regression when datasets contain heavy-tailed outliers or non-normal error distributions

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add choxos/biostatagent regression-models

For use in Claude.ai and ChatGPT

Key Features

01Performance optimization using QR decomposition for correlated predictors

02Expert recommendations for weakly informative prior distributions

03Specialized models for overdispersed counts and robust Student-t regression

040 GitHub stars

05Comprehensive templates for Linear, Logistic, and Poisson regression

06Support for Stan and JAGS probabilistic programming languages

Use Cases

01Building hierarchical or complex statistical models for biostatistics and academic research

02Converting frequentist models into a Bayesian framework for superior uncertainty quantification

03Implementing robust regression when datasets contain heavy-tailed outliers or non-normal error distributions