English

Estimating the concentration parameter of a von Mises distribution: a systematic simulation benchmark

Applications 2024-09-10 v1 Computation

Abstract

In directional statistics, the von Mises distribution is a key element in the analysis of circular data. While there is a general agreement regarding the estimation of its location parameter μ\mu, several methods have been proposed to estimate the concentration parameter κ\kappa. We here provide a thorough evaluation of the behavior of 12 such estimators for datasets of size NN ranging from 2 to 8\,192 generated with a κ\kappa ranging from 0 to 100. We provide detailed results as well as a global analysis of the results, showing that (1) for a given κ\kappa, most estimators have behaviors that are very similar for large datasets (N16N \geq 16) and more variable for small datasets, and (2) for a given estimator, results are very similar if we consider the mean absolute error for κ1\kappa \leq 1 and the mean relative absolute error for κ1\kappa \geq 1.

Keywords

Cite

@article{arxiv.2111.09660,
  title  = {Estimating the concentration parameter of a von Mises distribution: a systematic simulation benchmark},
  author = {Guillaume Marrelec and Alain Giron},
  journal= {arXiv preprint arXiv:2111.09660},
  year   = {2024}
}
R2 v1 2026-06-24T07:43:25.615Z