C-statistic vs bayesian : a comparative analysis of x-ray spectral statistics

Abstract

The X-ray spectra of active galactic nuclei (AGN) contain many spectral features, such as emission lines, that provide much insight into the inner workings of these objects. In a class of AGN known as Narrow-Line Seyfert 1 (NLS1) galaxies, analyzing these spectral features can provide especially useful insight into the internal structure due to their orientation. Using two different statistical methods (Bayesian X-ray Analysis and C-Statistic) to analyze the spectra of 21 NLS1 galaxies, it can be determined which method is superior at detecting emission lines in the spectra. It is found that C-Statistic may be superior at detecting lines in low-count spectra, but BXA has separate advantages that may lead to more success at detecting lines in real sources.