The problem of parametric, autoregressive model based estimation of a time-varying spectral density function of a multivariate nonstationary process is considered. It is shown that estimation results can be considerably improved if identification of the autoregressive model is carried out using the two-sided doubly exponentially weighted lattice algorithm which combines results yielded by two one-sided lattice algorithms running...

The problem of parametric, autoregressive model based estimation of a time-varying spectral density function of a nonstationary process is considered. It is shown that estimation results can be considerably improved if identification of the autoregressive model is carried out using the two-sided doubly exponentially weighted lattice algorithm which combines results yielded by two one-sided lattice algorithms running forward in...

Autoregressive modeling is a widespread parametricspectrum estimation method. It is well known that, in the caseof stationary processes with unknown order, its accuracy canbe improved by averaging models of different complexity usingsuitably chosen weights. The paper proposes an extension of thistechnique to the case of multivariate locally stationary processes.The proposed solution is based on local autoregressive...