Finite-sample bias-correction factors for the median absolute deviation based on the Harrell-Davis quantile estimator and its trimmed modification

The repository contains auxiliary data for the paper "Finite-sample bias-correction factors for the median absolute deviation based on the Harrell-Davis quantile estimator and its trimmed modification":

• The source code of the paper: mad-factors.Rmd, preamble.tex, references.bib, *.R
• Simulation 1 "Evaluating bias-correction factors using the Monte-Carlo method":
  • Source code: simulation-factors/
  • The results: data/sm.csv, data/hd.csv, data/thd-sqrt.csv
• Simulation 2 "Statistical efficiency of the median absolute deviation":
  • Source code: simulation-efficiency.R
  • The results: data/efficiency.csv
• Simulation 3 "Sensitivity to outliers of the median absolute deviation"
  • Source code: simulation-sensitivity.R
  • The results: data/sensitivity-sd.csv, data/sensitivity-iqr.csv, data/sensitivity-mad.csv
• The reference implementation of the suggested estimators: reference-implementation.R