stat-analysis

Execute reproducible statistical analyses from research plans to result reports.

8|Updated Mar 18, 2026
One-click install
npx skills add https://github.com/TerryFYL/ai-research-army --skill stat-analysis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: stat-analysis
Source: https://github.com/TerryFYL/ai-research-army/tree/main/skills/stat-analysis
Command: npx skills add https://github.com/TerryFYL/ai-research-army --skill stat-analysis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates end-to-end statistical analyses aligned with a research plan, enabling hypothesis-driven, reproducible workflows from design through results.

Core Features & Use Cases

  • Environment preparation & plan parsing: Reads the research_plan.md and data dictionaries, sets up a reproducible analysis workspace.
  • Hypothesis declaration & prereq checks: Explicitly defines hypotheses and validates statistical assumptions before modeling.
  • Multi-path analysis & model comparison: Runs multiple feasible methods and compares them using fit metrics and interpretability criteria.
  • Baseline, primary, subgroup, and sensitivity analyses: Generates a comprehensive workflow including Table 1, main results, subgroup analyses, and sensitivity checks.
  • Reproducible deliverables: Outputs analysis_results.md and an analysis codebase (analysis/) for full traceability and replication.

Quick Start

Describe your research plan and dataset, then run the full statistical analysis from preprocessing to results delivery.

Quick Start

Describe your research plan and dataset, then run the full statistical analysis from preprocessing to results delivery.

Frequently Asked Questions about stat-analysis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run reproducible statistical analyses from a research plan?

To run reproducible statistical analyses, you provide a data-ready dataset and a research_plan.md, and the workflow parses your design to generate fully traceable analysis code and results from hypothesis specification to delivery.

What is included in an end-to-end statistical analysis workflow?

An end-to-end statistical analysis workflow includes environment preparation, hypothesis declaration, prerequisite checks, multi-path model comparison, and generates baseline tables, primary analyses, subgroup analyses, and sensitivity checks.

Can I compare multiple statistical methods and generate baseline tables automatically?

Yes, you can compare multiple feasible statistical methods using fit metrics and interpretability criteria, while automatically generating comprehensive baseline tables and sensitivity checks alongside primary analyses.

Do I need pandas and statsmodels to perform hypothesis-driven data analysis?

Yes, you need access to Python or R toolchains including pandas, statsmodels, and lifelines to execute the hypothesis-driven data analysis and generate the reproducible analysis codebase.

What's the best way to ensure traceability in research data analysis projects?

The best way to ensure traceability in research data analysis is to output an analysis_results.md file alongside a version-controlled analysis codebase, providing full replication from the original research_plan.md.