Multivariate analysis of BATSE gamma-ray burst properties using skewed distributions

TL;DR

This study applies skewed distributions to multivariate analysis of BATSE gamma-ray burst data, finding that the number of classes varies and some identified classes may be artifacts of data limitations.

Contribution

It introduces a comprehensive multivariate analysis using skewed distributions, revealing variability in the number of GRB classes and highlighting potential artifacts.

Findings

Number of classes varies from 2 to 4 depending on parameter space

Higher-dimensional spaces tend to identify more classes

Monte Carlo tests suggest some classes may be spurious

Abstract

The number of classes of gamma-ray bursts (GRBs), besides the well-established short and long ones, remains a debatable issue. It was already shown, however, that when invoking skewed distributions, the $\log T_{90}$ and $\log T_{90} - \log H_{32}$ spaces are adequately modeled with mixtures of only two such components, implying two GRB types. Herein, a comprehensive multivariate analysis of several multi-dimensional parameter spaces is conducted for the BATSE sample of GRBs, with the usage of skewed distributions. It is found that the number of extracted components varies between the examined parameter sets, and ranges from 2 to 4, with higher-dimensional spaces allowing for more classes. A Monte Carlo testing implies that these additional components are likely to be artifacts owing to the finiteness of the data and be a result of examining a particular realization of the data as a random sample, resulting in spurious identifications.