Parameter estimation for pure diagonal bilinear time series: An algorithm for maximum likelihood procedure

by Bouzaachane Khadija, Harti Mostafa, Benghabrit Youssef .

Abstract: In this work, a new estimate algorithm for the parameters of a pure diagonal bilinear model is proposed. This algorithm turns out to be very reliable in estimating the true parameters's values of a given model. It combines maximum likelihood method, Kalman filter algorithm and simulated annealing. Simulation results demonstrate that the algorithm is succeeds and promising.

Key Words: ime Series, Pure Diagonal Bilinear Model, Maximum LikeLihood, Kalman Filter, Simulated Annealing

Authors:
Bouzaachane Khadija, bouzaachane@netcourrier.com
Harti Mostafa, mharti@fsdmfes.ac.ma
Benghabrit Youssef, you_benghabrit@yahoo.fr

Editor: geel, Mohammed, MIAQEEL@kku.edu.sa

READING THE ARTICLE: You can read the article in portable document (.pdf) format (126683 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 2284 times since October 3, 2007.


Return to the InterStat Home Page.