Computational aspects of the optimal transit path problem

In this paper we present a novel method for short term forecast of time series based on Knot-Optimizing Spline Networks (KOSNETS). The time series is first approximated by a nonlinear recurrent system. The resulting recurrent system is then approximated by feedforward B-spline networks, yielding a n...

Full description

Bibliographic Details
Main Authors: Caccetta, Louis, Loosen, Ian, Rehbock, Volker
Format: Journal Article
Published: American Institute of Mathematical Sciences AIMS 2008
Online Access:http://aimsciences.org/journals/jimo/contents.jsp
http://hdl.handle.net/20.500.11937/3053
_version_ 1848744124847489024
author Caccetta, Louis
Loosen, Ian
Rehbock, Volker
author_facet Caccetta, Louis
Loosen, Ian
Rehbock, Volker
author_sort Caccetta, Louis
building Curtin Institutional Repository
collection Online Access
description In this paper we present a novel method for short term forecast of time series based on Knot-Optimizing Spline Networks (KOSNETS). The time series is first approximated by a nonlinear recurrent system. The resulting recurrent system is then approximated by feedforward B-spline networks, yielding a nonlinear optimization problem. In this optimization problem, both the knot points and the coefficients of the B-splines are decision variables so that the solution to the problem has both optimal coefficients and partition points. To demonstrate the usefulness and accuracy of the method, numerical simulations and tests using various model and real time series are performed. The numerical simulation results are compared with those from a well-known regression method, MARS. The comparison shows that our method outperforms MARS for nonlinear problems.
first_indexed 2025-11-14T05:56:29Z
format Journal Article
id curtin-20.500.11937-3053
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T05:56:29Z
publishDate 2008
publisher American Institute of Mathematical Sciences AIMS
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-30532017-01-30T10:28:16Z Computational aspects of the optimal transit path problem Caccetta, Louis Loosen, Ian Rehbock, Volker In this paper we present a novel method for short term forecast of time series based on Knot-Optimizing Spline Networks (KOSNETS). The time series is first approximated by a nonlinear recurrent system. The resulting recurrent system is then approximated by feedforward B-spline networks, yielding a nonlinear optimization problem. In this optimization problem, both the knot points and the coefficients of the B-splines are decision variables so that the solution to the problem has both optimal coefficients and partition points. To demonstrate the usefulness and accuracy of the method, numerical simulations and tests using various model and real time series are performed. The numerical simulation results are compared with those from a well-known regression method, MARS. The comparison shows that our method outperforms MARS for nonlinear problems. 2008 Journal Article http://hdl.handle.net/20.500.11937/3053 http://aimsciences.org/journals/jimo/contents.jsp American Institute of Mathematical Sciences AIMS fulltext
spellingShingle Caccetta, Louis
Loosen, Ian
Rehbock, Volker
Computational aspects of the optimal transit path problem
title Computational aspects of the optimal transit path problem
title_full Computational aspects of the optimal transit path problem
title_fullStr Computational aspects of the optimal transit path problem
title_full_unstemmed Computational aspects of the optimal transit path problem
title_short Computational aspects of the optimal transit path problem
title_sort computational aspects of the optimal transit path problem
url http://aimsciences.org/journals/jimo/contents.jsp
http://hdl.handle.net/20.500.11937/3053