Getting Started with scanr

1 Overview

The scanr package detects multiple change points in univariate time series using window-based SCAN statistics. The method compares local windows along a series, calibrates evidence for changes using resampling, and combines information across several window sizes.

This book demonstrates a typical workflow:

  1. simulate a univariate time series with known change points;
  2. run scan_cpd() with multiple window sizes;
  3. compare estimated change points with the known truth;
  4. inspect the result using diagnostic visualizations;
  5. repeat the workflow for distributional changes;
  6. use the SWAL statistic for single change-point localization.