Paper Title:

A Novel Method to Estimate ARMA Time-Varying Parameter Using Grey Forecasting Model

Periodical Advanced Materials Research (Volumes 532 - 533)
Main Theme Materials Science and Information Technology II
Edited by Suozhang Cai and Mingli Li
Pages 876-880
DOI 10.4028/www.scientific.net/AMR.532-533.876
Citation Wei Zhen Chen, 2012, Advanced Materials Research, 532-533, 876
Online since June, 2012
Authors Wei Zhen Chen
Keywords ARMA Model, Grey Forecasting Model, Model Parameter, Time-Varying System
Price US$ 28,-
Article Preview
View full size
Abstract

In this paper, a novel method to estimate ARMA time-varying parameter is proposed, based on ESCGMmv grey forecasting model. The estimation procedure of this method consists of two phases, first estimating online ARMA parameter by the least squares method. Then, using ESCGMmv grey forecasting model to forecast the error between and .The model gives an adequate description of the system under consideration. Finally, modify the estimate parameter by the estimate error. Simulation results confirm that the propose method can obtain more accurate estimate time- varying parameter.