Title :
Linear estimation of ARMA processes
Author :
Moura, José M F ; Ribeiro, M.I.
Author_Institution :
Dept. of Electr. & Comput. Eng., Carnegie-Mellon Univ., Pittsburgh, PA, USA
Abstract :
The authors propose an identification algorithm for autoregressive moving average (ARMA) processes. Given a finite length sample drawn from an ARMA (p0, q0) model, the technique provides the estimated values of the orders p0 and p0, as well as the AR and MA coefficients. They are obtained from the reflection coefficient sequence estimated directly from the data. The order selection scheme is based upon the minimization of a functional that measures the mismatch of the data to any ARMA (p,q) assumed as its model
Keywords :
identification; minimisation; parameter estimation; spectral analysis; statistics; ARMA processes; autoregressive moving average; finite length sample; functional; identification algorithm; linear estimation; minimization; mismatch; order selection scheme; reflection coefficient sequence; Autocorrelation; Ducts; Ear; Equations; Linearity; Maximum likelihood estimation; Optimization methods; Poles and zeros; Power system modeling; Reflection;
Conference_Titel :
Spectrum Estimation and Modeling, 1988., Fourth Annual ASSP Workshop on
Conference_Location :
Minneapolis, MN
DOI :
10.1109/SPECT.1988.206197
