Extension of autocovariance coefficients sequence for periodically correlated random processes by using the partial autocorrelation function

The extension of stationary process autocorrelation coefficients sequence is a classical problem in the field of spectral estimation. The periodically correlated (PC) processes have praticai importance and an interest according to their connection with stationary multivariate processes. That's why we propose a new approach to resolve the previous problem in this context. We use the partial autocorrelation function (PACF) of this processes class. The extension is so easy to describe. Next, we extend the maximum entropy method (MEM) to the degenerate case and show that the solution is given by a Periodic Autoregressive (PAR) process. Furthermore, the connection with the problem of multivariate stationary processes autocorrelation sequence is presented.