Bayesian estimation of correlation functions

We apply Bayesian statistics to the estimation of correlation functions. give probability distributions auto- and cross-correlations as functions data. Our procedure uses measured data optimally informs about certainty level estimation. results general stationary processes their essence is a nonparametric spectra. It allows one better understand statistical noise fluctuations, assess correlations between two variables, postulate parametric models spectra that can be further tested. also propose method numerically generate correlated with given spectrum.