Comparison between adaptive and non-adaptive HRBF neural network in multiple steps time series forecasting

Mazlina Mamat, Salina Abdul Samad

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

This paper compares the performance of adaptive and non-adaptive learning approaches of the Hybrid Radial Basis Function (HRBF) neural network in multiple steps time series forecasting. The HRBF was trained by using the Adaptive Fuzzy C-Means Clustering (AFCMC) and Exponential Weighted Recursive Least Square (e-WRLS) algorithms. Both approaches were set to produce up to 25 steps ahead forecasting on two time series data: Mackey Glass and Set A Data from Santa Fe Competition. The performance of both approaches in multiple steps ahead forecasting was measured using Mean Square Error Test and Coefficient of Determination Test between the actual and forecasted data for the 25 steps ahead forecasting. Results show that both approaches perform comparatively equal for shorter forecasting distance. However for longer forecasting distance (10 steps ahead onwards), the adaptive approach performs significantly better to compare with non-adaptive approach.

Original languageEnglish
Title of host publication2nd International Conference on Computer Research and Development, ICCRD 2010
Pages817-821
Number of pages5
DOIs
Publication statusPublished - 2010
Event2nd International Conference on Computer Research and Development, ICCRD 2010 - Kuala Lumpur
Duration: 7 May 201010 May 2010

Other

Other2nd International Conference on Computer Research and Development, ICCRD 2010
CityKuala Lumpur
Period7/5/1010/5/10

Fingerprint

Time series
Neural networks
Mean square error
Glass

Keywords

  • Adaptive
  • HRBF
  • Non-adaptive
  • Time series forecasting

ASJC Scopus subject areas

  • Computer Science (miscellaneous)

Cite this

Mamat, M., & Abdul Samad, S. (2010). Comparison between adaptive and non-adaptive HRBF neural network in multiple steps time series forecasting. In 2nd International Conference on Computer Research and Development, ICCRD 2010 (pp. 817-821). [5489479] https://doi.org/10.1109/ICCRD.2010.177

Comparison between adaptive and non-adaptive HRBF neural network in multiple steps time series forecasting. / Mamat, Mazlina; Abdul Samad, Salina.

2nd International Conference on Computer Research and Development, ICCRD 2010. 2010. p. 817-821 5489479.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Mamat, M & Abdul Samad, S 2010, Comparison between adaptive and non-adaptive HRBF neural network in multiple steps time series forecasting. in 2nd International Conference on Computer Research and Development, ICCRD 2010., 5489479, pp. 817-821, 2nd International Conference on Computer Research and Development, ICCRD 2010, Kuala Lumpur, 7/5/10. https://doi.org/10.1109/ICCRD.2010.177
Mamat M, Abdul Samad S. Comparison between adaptive and non-adaptive HRBF neural network in multiple steps time series forecasting. In 2nd International Conference on Computer Research and Development, ICCRD 2010. 2010. p. 817-821. 5489479 https://doi.org/10.1109/ICCRD.2010.177
Mamat, Mazlina ; Abdul Samad, Salina. / Comparison between adaptive and non-adaptive HRBF neural network in multiple steps time series forecasting. 2nd International Conference on Computer Research and Development, ICCRD 2010. 2010. pp. 817-821
@inproceedings{118b14d97fd5448d84b985d80a7413f2,
title = "Comparison between adaptive and non-adaptive HRBF neural network in multiple steps time series forecasting",
abstract = "This paper compares the performance of adaptive and non-adaptive learning approaches of the Hybrid Radial Basis Function (HRBF) neural network in multiple steps time series forecasting. The HRBF was trained by using the Adaptive Fuzzy C-Means Clustering (AFCMC) and Exponential Weighted Recursive Least Square (e-WRLS) algorithms. Both approaches were set to produce up to 25 steps ahead forecasting on two time series data: Mackey Glass and Set A Data from Santa Fe Competition. The performance of both approaches in multiple steps ahead forecasting was measured using Mean Square Error Test and Coefficient of Determination Test between the actual and forecasted data for the 25 steps ahead forecasting. Results show that both approaches perform comparatively equal for shorter forecasting distance. However for longer forecasting distance (10 steps ahead onwards), the adaptive approach performs significantly better to compare with non-adaptive approach.",
keywords = "Adaptive, HRBF, Non-adaptive, Time series forecasting",
author = "Mazlina Mamat and {Abdul Samad}, Salina",
year = "2010",
doi = "10.1109/ICCRD.2010.177",
language = "English",
isbn = "9780769540436",
pages = "817--821",
booktitle = "2nd International Conference on Computer Research and Development, ICCRD 2010",

}

TY - GEN

T1 - Comparison between adaptive and non-adaptive HRBF neural network in multiple steps time series forecasting

AU - Mamat, Mazlina

AU - Abdul Samad, Salina

PY - 2010

Y1 - 2010

N2 - This paper compares the performance of adaptive and non-adaptive learning approaches of the Hybrid Radial Basis Function (HRBF) neural network in multiple steps time series forecasting. The HRBF was trained by using the Adaptive Fuzzy C-Means Clustering (AFCMC) and Exponential Weighted Recursive Least Square (e-WRLS) algorithms. Both approaches were set to produce up to 25 steps ahead forecasting on two time series data: Mackey Glass and Set A Data from Santa Fe Competition. The performance of both approaches in multiple steps ahead forecasting was measured using Mean Square Error Test and Coefficient of Determination Test between the actual and forecasted data for the 25 steps ahead forecasting. Results show that both approaches perform comparatively equal for shorter forecasting distance. However for longer forecasting distance (10 steps ahead onwards), the adaptive approach performs significantly better to compare with non-adaptive approach.

AB - This paper compares the performance of adaptive and non-adaptive learning approaches of the Hybrid Radial Basis Function (HRBF) neural network in multiple steps time series forecasting. The HRBF was trained by using the Adaptive Fuzzy C-Means Clustering (AFCMC) and Exponential Weighted Recursive Least Square (e-WRLS) algorithms. Both approaches were set to produce up to 25 steps ahead forecasting on two time series data: Mackey Glass and Set A Data from Santa Fe Competition. The performance of both approaches in multiple steps ahead forecasting was measured using Mean Square Error Test and Coefficient of Determination Test between the actual and forecasted data for the 25 steps ahead forecasting. Results show that both approaches perform comparatively equal for shorter forecasting distance. However for longer forecasting distance (10 steps ahead onwards), the adaptive approach performs significantly better to compare with non-adaptive approach.

KW - Adaptive

KW - HRBF

KW - Non-adaptive

KW - Time series forecasting

UR - http://www.scopus.com/inward/record.url?scp=77955194793&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955194793&partnerID=8YFLogxK

U2 - 10.1109/ICCRD.2010.177

DO - 10.1109/ICCRD.2010.177

M3 - Conference contribution

SN - 9780769540436

SP - 817

EP - 821

BT - 2nd International Conference on Computer Research and Development, ICCRD 2010

ER -