On the estimation of non-stationary functional series tarma models

Author

Poulimenos, Aggelos G. ; Fassois, Spilios D.

Author_Institution

Dept. of Mech. & Aeronaut. Eng., Univ. of Patras, Patras, Greece

fYear

2005

fDate

4-8 Sept. 2005

Firstpage

1

Lastpage

4

Abstract

Maximum Likelihood (ML) and Multi-Stage Weighted Linear / Non-Linear Least Squares (MS-WLLS / MS-WNLS) estimation methods are formulated for non-stationary Functional Series Time-dependent ARMA (FS-TARMA) models. The methods´ effectiveness, as well as their superiority over Ordinary Linear / Non-Linear Least Squares (OLLS / ONLS) type methods not accounting for innovations serial heteroscedasticity, are demonstrated via Monte Carlo experiments.

Keywords

autoregressive moving average processes; least squares approximations; maximum likelihood estimation; signal processing; MS-WLLS-MS-WNLS; Monte Carlo experiments; OLLS-ONLS type methods; innovation serial heteroscedasticity; maximum likelihood estimation methods; multistage weighted linear estimation methods; nonlinear least square estimation methods; nonstationary functional series TARMA model estimation; ordinary linear-nonlinear least squares type methods; time-dependent ARMA models; Autoregressive processes; Least squares approximations; Maximum likelihood estimation; Standards; Technological innovation; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing Conference, 2005 13th European

Conference_Location

Antalya

Print_ISBN

978-160-4238-21-1

Type

conf

Filename

7078497