Title: Evolutionary algorithm for self-exciting threshold autoregressive model

Authors: Jian Hua Gao; Yuan Xiang Li; Hua Jiang; Wen Yong Dong

Addresses: School of Computer, Wuhan University, Wuhan Hubei, China ' School of Computer, Wuhan University, Wuhan Hubei, China ' School of Computer, Wuhan University, Wuhan Hubei, China ' School of Computer, Wuhan University, Wuhan Hubei, China

Abstract: This paper proposes an evolutionary algorithm for getting the parameters of the self-exciting threshold autoregressive models. The algorithm aims to consider the number of the threshold intervals as endogenous variables and involves the threshold argument vector and delay parameter simultaneously. The parameters of the AR models in each threshold interval will be evolved during the process of searching the threshold argument vector. The experiment shows that the model founded on our algorithm has good prediction results.

Keywords: self-exciting threshold autoregressive models; evolutionary algorithms; threshold argument vector; delay parameter; modelling.

DOI: 10.1504/IJWMC.2015.069389

International Journal of Wireless and Mobile Computing, 2015 Vol.8 No.3, pp.229 - 235

Received: 25 Jul 2014
Accepted: 03 Sep 2014

Published online: 14 May 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article