Multiple oscillation frequencies separated by close to or less than the formal frequency resolution of a data set are a serious problem in the frequency analysis of time series data. We present a new and fully automated Bayesian approach that searches for close frequencies in time series data and assesses their significance by comparison to no signal and a mono-periodic signal. We extensively test the approach with synthetic data sets and apply it to the 156 days-long high-precision BRITE photometry of the SPB star HD 201433, for which we find a sequence of nine statistically significant rotationally split dipole modes.
@article{arxiv.1611.08167,
title = {Bayesian frequency analysis of HD 201433 observations with BRITE},
author = {T. Kallinger and W. W. Weiss},
journal= {arXiv preprint arXiv:1611.08167},
year = {2016}
}
Comments
6 pages, will be published in the proceedings of the Second BRITE-Constellation Science Conference: small satellites - big science