OpenSkillIndex← back to search
claude-codescience

Babysitter robust-statistics-toolkit

Robust statistical methods resistant to outliers

install
source · Clone the upstream repo
git clone https://github.com/a5c-ai/babysitter
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/a5c-ai/babysitter "$T" && mkdir -p ~/.claude/skills && cp -r "$T/library/specializations/domains/science/mathematics/skills/robust-statistics-toolkit" ~/.claude/skills/a5c-ai-babysitter-robust-statistics-toolkit && rm -rf "$T"
manifest: library/specializations/domains/science/mathematics/skills/robust-statistics-toolkit/SKILL.md
tags
#statistical-analysis#outlier-detection#robust-regression#data-science#statistical-computing
source content

Robust Statistics Toolkit

Purpose

Provides robust statistical methods resistant to outliers and model violations for reliable inference.

Capabilities

  • M-estimators (Huber, Tukey)
  • Trimmed and winsorized estimators
  • Robust regression (MM-estimation)
  • Breakdown point analysis
  • Influence function computation
  • Robust covariance estimation

Usage Guidelines

  1. Outlier Detection: Identify potential outliers first
  2. Estimator Selection: Choose based on expected contamination
  3. Breakdown Point: Consider required breakdown point
  4. Efficiency: Balance robustness and efficiency

Tools/Libraries

  • robustbase (R)
  • scikit-learn
  • statsmodels