Can I perform Network Meta-Analysis (NMA) with this tool?

Yes, the skill includes consistency models for NMA to compare three or more treatments across different studies, including treatment ranking capabilities.

Does it account for publication bias?

Yes, the skill includes specialized selection models that adjust for the severity of publication bias by analyzing how z-scores impact the likelihood of publication.

How does it handle study heterogeneity?

It implements random effects models using non-centered parameterization and automatically calculates key statistics like I-squared and Tau-squared to quantify variation.

What modeling languages are supported by this skill?

This skill provides production-ready implementations for both Stan and JAGS, allowing flexibility based on your preferred probabilistic programming environment.

Bayesian Meta-Analysis

Name: Bayesian Meta-Analysis
Author: choxos

bychoxos

データサイエンスとML

Implements robust Bayesian meta-analysis models using Stan and JAGS to synthesize data across multiple studies with advanced statistical precision.

概要

This skill empowers researchers and data scientists to conduct sophisticated meta-analyses within the Claude environment. It provides a comprehensive library of Bayesian models ranging from simple fixed and random effects to complex network meta-analyses (NMA) and selection models for publication bias. By offering standardized implementations in both Stan and JAGS, it facilitates rigorous statistical synthesis of clinical trials, behavioral studies, or experimental data, ensuring best practices in prior selection and heterogeneity estimation are followed.

主な機能

Automated calculation of key statistics like I-squared and prediction intervals
Publication bias detection via selection model implementations
Specialized support for binary outcomes using log-odds ratios
Fixed and random effects modeling with non-centered parameterization
Comprehensive Network Meta-Analysis (NMA) for multi-treatment comparisons
0 GitHub stars

ユースケース

Comparing multiple treatments simultaneously through network meta-analysis
Synthesizing clinical trial results to determine overall treatment efficacy
Adjusting for potential publication bias in academic systematic reviews

概要

主な機能

Automated calculation of key statistics like I-squared and prediction intervals
Publication bias detection via selection model implementations
Specialized support for binary outcomes using log-odds ratios
Fixed and random effects modeling with non-centered parameterization
Comprehensive Network Meta-Analysis (NMA) for multi-treatment comparisons
0 GitHub stars

ユースケース

Comparing multiple treatments simultaneously through network meta-analysis
Synthesizing clinical trial results to determine overall treatment efficacy
Adjusting for potential publication bias in academic systematic reviews