OpenSkillIndex← back to search
claude-codescience

Babysitter iminuit-statistical-fitter

iminuit statistical fitting skill for physics data analysis with proper error handling and profile likelihood

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 "$T/library/specializations/domains/science/physics/skills/iminuit-statistical-fitter" ~/.claude/skills/a5c-ai-babysitter-iminuit-statistical-fitter && rm -rf "$T"
manifest: library/specializations/domains/science/physics/skills/iminuit-statistical-fitter/SKILL.md
tags
#physics-fitting#statistical-analysis#error-handling#profile-likelihood#iminuit
source content

iminuit Statistical Fitter

Purpose

Provides expert guidance on iminuit for statistical fitting in physics, including proper error estimation and profile likelihood calculations.

Capabilities

  • MINUIT minimization algorithms
  • HESSE error matrix calculation
  • MINOS asymmetric error estimation
  • Profile likelihood computation
  • Constrained fitting
  • Simultaneous fit orchestration

Usage Guidelines

  1. Model Definition: Define cost function for minimization
  2. Minimization: Run MIGRAD for parameter estimation
  3. Error Analysis: Use HESSE and MINOS for uncertainties
  4. Profile Likelihood: Compute profile likelihood for parameters
  5. Simultaneous Fits: Combine multiple datasets in fits

Tools/Libraries

  • iminuit
  • probfit
  • zfit