Parameter estimation for compact binary inspirals with a simple noise realization

11th Edoardo Amaldi Conference on Gravitational Waves

Kim, Jeongcho^1 ; Kim, Chunglee^2 ; Lee, Hyung Won^1

Department of Computer Simulation, Inje University, Gimhae

50834, Korea, Republic of^1

Yonsei University Observatory, Seoul

03722, Korea, Republic of^2

关键词: Detector noise;    Gaussians;    Markov chain Monte Carlo;    Most likely;    Noise realizations;    Non-Gaussian;    Specific frequencies;    Stationary noise;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/716/1/012033/pdf
DOI  :  10.1088/1742-6596/716/1/012033

来源: IOP

【 摘 要 】

In the context of parameter estimation of gravitational waves (GWs), detector noise is assumed to be Gaussian and stationary. In reality, many electric glitches, which are neither Gaussian nor stationary, were observed and reported in publications by the LSC-Virgo collabotation. Proper noise reduction is important in GW data analysis, as these glitches would limit, if not downgrade, the quality of parameter estimation. In this work, we investigate the accuracy of results obtained by Markov Chain Monte Carlo (MCMC) parameter estimation for compact binary inspirals with the LIGO-Virgo network when non-Gaussian, stationary noise is remained in data of each interferometer. Spiky, delta function-like glitches, which are stationary, do not affect correlations between parameters. However, most likely values of chirp mass and distance seem to be shifted by the specific frequencies and amplitudes of glitches.