Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal MATHunesa: Jurnal Ilmiah Matematika
Ellyana, Elfi
Unknown Affiliation
Mathematics

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
PERBANDINGAN METODE VECTOR AUTOREGRESSIVE NEURAL NETWORK (VAR-NN) DAN ELMAN RECURRENT NEURAL NETWORK (ERNN) UNTUK PERAMALAN JUMLAH PENUMPANG KERETA API Ellyana, Elfi
MATHunesa: Jurnal Ilmiah Matematika Vol. 13 No. 1 (2025)
Publisher : Universitas Negeri Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/mathunesa.v13n1.p125-138

Abstract

Vector Autoregressive Neural Network (VAR-NN) is a combination of VAR and Neural Network that has the potential to improve forecasting accuracy especially in the case of data that has significant non-linear patterns while Elman Recurrent Neural Network (ERNN) is effective in recognizing non-linear patterns in complex time series data. This study aims to determine the best modeling of VAR-NN and ERNN for forecasting the number of train passengers in Java (Jabotabek and Non Jabotabek). The results showed that the best VAR-NN model for the Jabotabek area is the VAR-NN (1-7-1) model with MSE and MAPE test values of 0.0137 and 11.7% and for the Non Jabotabek area the VAR-NN (2-14-1) model with MSE and MAPE test values of 0.0165 and 21%, respectively, 0165 and 21%, while the best ERNN model for the Jabotabek area is the ERNN (5-15-1) model with MSE and MAPE values of 3.4983e+07 and 38.7995% and for the Non Jabotabek area is the ERNN (6-15-1) model with MSE and MAPE values of 3.4591e+06 and 50.8854%. This study concludes that the best model for forecasting the number of train passengers in the Jabotabek area is the VAR-NN (1-7-1) model and the Non Jabotabek area is the VAR-NN (2-14-1) model.
Co-Authors