Abstract
A procedure is presented for generating an ARMA spectral model of a stationary time series based upon a finite set of time series observations. The ARMA model's coefficients are estimated by utilizing a basic difference equation characterizing the underlying rational spectral model. In examples treated to date, this new procedure has been found to produce "significantly" better spectral estimates when compared to the maximum entropy method and other AR and ARMA based procedures. These examples have included both narrow and broadband spectra. With this in mind, the new ARMA procedure promises to be an important spectral estimation tool.

This publication has 4 references indexed in Scilit: