eprintid: 1640
rev_number: 7
eprint_status: archive
userid: 6
dir: disk0/00/00/16/40
datestamp: 2013-07-10 10:42:01
lastmod: 2013-07-10 10:42:01
status_changed: 2013-07-10 10:42:01
type: article
metadata_visibility: show
creators_name: Giovanis, Eleftherios
creators_id: eleftherios.giovanis@imtlucca.it
title: Ordinary Least Squares and Genetic Algorithms Optimization in Smoothing Transition Autoregressive (STAR) Models
ispublished: pub
subjects: HA
subjects: HB
divisions: EIC
full_text_status: none
keywords: Genetic Algorithms; MATLAB; Membership Functions; Smoothing Transition Autoregressive
abstract: Abstract—In this paper we present, propose and examine
additional membership functions as also we propose least squares with genetic algorithms optimization in order to find the optimum fuzzy membership functions parameters. More
specifically, we present the tangent hyperbolic, Gaussian and Generalized bell functions. The reason we propose that is because Smoothing Transition Autoregressive (STAR) models follow fuzzy logic approach therefore more functions should be tested. Some numerical applications for S&P 500, FTSE 100 stock returns and for unemployment rate are presented and MATLAB routines are provided
date: 2011-03
date_type: published
publication: International Journal of Computer Information Systems
volume: 2
number: 3
publisher: Silicon Valley Publishers
pagerange: 17-23
refereed: TRUE
issn: 2229-5208
official_url: http://www.svpublishers.co.uk/#/ijcis-mar-2011/4550826231
citation:   Giovanis, Eleftherios  Ordinary Least Squares and Genetic Algorithms Optimization in Smoothing Transition Autoregressive (STAR) Models.  International Journal of Computer Information Systems, 2 (3).  pp. 17-23.  ISSN 2229-5208      (2011)