English

Custom Orthogonal Weight functions (COWs) for Event Classification

Methodology 2022-08-24 v2 High Energy Physics - Experiment Data Analysis, Statistics and Probability

Abstract

A common problem in data analysis is the separation of signal and background. We revisit and generalise the so-called sWeightssWeights method, which allows one to calculate an empirical estimate of the signal density of a control variable using a fit of a mixed signal and background model to a discriminating variable. We show that sWeightssWeights are a special case of a larger class of Custom Orthogonal Weight functions (COWs), which can be applied to a more general class of problems in which the discriminating and control variables are not necessarily independent and still achieve close to optimal performance. We also investigate the properties of parameters estimated from fits of statistical models to sWeightssWeights and provide closed formulas for the asymptotic covariance matrix of the fitted parameters. To illustrate our findings, we discuss several practical applications of these techniques.

Keywords

Cite

@article{arxiv.2112.04574,
  title  = {Custom Orthogonal Weight functions (COWs) for Event Classification},
  author = {Hans Dembinski and Matthew Kenzie and Christoph Langenbruch and Michael Schmelling},
  journal= {arXiv preprint arXiv:2112.04574},
  year   = {2022}
}

Comments

18 pages, 16 figures, for associated software tools see https://pypi.org/project/sweights/

R2 v1 2026-06-24T08:09:48.079Z