On Affecting the Frequency Domain Model Error Distribution in Autoregressive Spectral Estimation using Prefiltering


This paper treats the problem of approximating a complex stochastic process in a given frequency region by an estimated autoregressive (AR) model. A standard approach to affect the model error distribution in identification is the use of prefiltering. For example, if the process can be described by an ARMA model with known zeros, the maximum likelihood… (More)

8 Figures and Tables


  • Presentations referencing similar topics