Bayesian estimation of correlation functions
Abstract
We apply Bayesian statistics to the estimation of correlation functions. We give the probability distributions of auto- and cross-correlations as functions of the data. Our procedure uses the measured data optimally and informs about the certainty level of the estimation. Our results apply to general stationary processes and their essence is a non-parametric estimation of spectra. It allows one to better understand the statistical noise fluctuations, assess the correlations between two variables, and postulate parametric models of spectra that can be further tested. We also propose a method to numerically generate correlated noise with a given spectrum.
Cite
@article{arxiv.2205.03611,
title = {Bayesian estimation of correlation functions},
author = {Angel Gutierrez-Rubio and Juan S. Rojas-Arias and Jun Yoneda and Seigo Tarucha and Daniel Loss and Peter Stano},
journal= {arXiv preprint arXiv:2205.03611},
year = {2022}
}
Comments
See (v1) for the original Angel's version and the article's history background. Changes: different priors for zero frequency; Fig. 5 redrawn; Section VI renamed and extended; App. B extended by non-factorizable priors; Added Apps. D, E, F, G, and H, and 9 references; Style and format changes throughout. Many of these changes reflect a year of experience in applying the theory to real-world data