Can it perform Bayesian alternatives to frequentist tests?

Yes, it leverages PyMC and ArviZ to provide Bayesian versions of t-tests and ANOVA, offering Bayes Factors and posterior probability statements.

Does this skill support non-parametric tests?

Yes, it includes robust alternatives such as Mann-Whitney U, Wilcoxon signed-rank, and Kruskal-Wallis tests for datasets that violate normality assumptions.

What Python libraries are required for this skill?

The skill utilizes industry-standard libraries including Scipy.stats, Statsmodels, Pingouin for general statistics, and PyMC for Bayesian modeling.

Does it check for statistical assumptions automatically?

Yes, it includes systematic verification for normality, homogeneity of variance, outliers, and linearity before running tests to ensure results are valid.

Can I use this for professional APA reporting?

Absolutely. The skill is specifically designed to generate statistical results in standard APA format, including effect sizes and confidence intervals, suitable for research papers.

Statistical Analysis Toolset

Name: Statistical Analysis Toolset
Author: lifangda

bylifangda

•

データサイエンスとML

Performs comprehensive hypothesis testing, regression analysis, and Bayesian statistics with automated assumption checking and APA-style reporting.

概要

This skill provides a robust framework for academic-grade statistical analysis within Claude Code, bridging the gap between raw data and publication-ready results. It streamlines the entire research workflow—from a priori power analysis and test selection to rigorous assumption verification and professional reporting. By integrating standard Python libraries like Scipy, Statsmodels, and Pingouin, it ensures that researchers can execute complex analyses—including ANOVA, logistic regression, and Bayesian modeling—with diagnostic visualizations and standardized APA output that meets rigorous academic standards.

主な機能

Automated assumption checking with diagnostic visualizations like Q-Q plots and residual analysis.
16 GitHub stars
Comprehensive hypothesis testing suite including t-tests, ANOVA, Chi-square, and non-parametric alternatives.
Bayesian statistical modeling using PyMC and ArviZ for robust probability estimation and Bayes Factors.
Standardized APA-style reporting and publication-ready data visualization generation.
Advanced regression modeling (linear, multiple, logistic) with VIF multicollinearity checks.

ユースケース

Conducting power analyses to determine required sample sizes during the study planning phase.
Analyzing experimental research data for academic publication or thesis work.
Automating the diagnostic check and assumption verification process for large datasets to ensure model validity.

概要

主な機能

Automated assumption checking with diagnostic visualizations like Q-Q plots and residual analysis.
16 GitHub stars
Comprehensive hypothesis testing suite including t-tests, ANOVA, Chi-square, and non-parametric alternatives.
Bayesian statistical modeling using PyMC and ArviZ for robust probability estimation and Bayes Factors.
Standardized APA-style reporting and publication-ready data visualization generation.
Advanced regression modeling (linear, multiple, logistic) with VIF multicollinearity checks.

ユースケース

Conducting power analyses to determine required sample sizes during the study planning phase.
Analyzing experimental research data for academic publication or thesis work.
Automating the diagnostic check and assumption verification process for large datasets to ensure model validity.