Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Zero : Jurnal Sains, Matematika, dan Terapan
Hari Wijayanto
Study Program of Statistics and Data Science, IPB University, Bogor, Indonesia
Agriculture, Biological Sciences & Forestry Chemistry Mathematics Physics

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

Found 1 Documents
Search

 1 
Robustness Evaluation of ANFIS, Hybrid GA-SVM, and SVM under Controlled Time Series Structures Sri Kustiara; Kusman Sadik; Hari Wijayanto
ZERO: Jurnal Sains, Matematika dan Terapan Vol 10, No 1 (2026): Zero: Jurnal Sains Matematika dan Terapan
Publisher : UIN Sumatera Utara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30829/zero.v10i1.29313

Abstract

This study evaluates the robustness of ANFIS, hybrid GA-SVM, and SVM under synthetic time-series structures using a factorial simulation framework combined with empirical validation. From a practical perspective, robust coal price forecasting is essential for supporting energy planning, trade management, and policy decision-making under uncertain market conditions. Empirical analysis of Indonesian coal prices reveals nonstationary behaviour, high volatility, and nonlinear dynamics. Forecasting performance is assessed using walk-forward validation, where SVM and hybrid GA-SVM demonstrate comparable accuracy and outperform ANFIS on the empirical dataset. To systematically examine model sensitivity to structural variations, a  factorial simulation design is implemented by varying seasonality, volatility, and predictor–response structure across 12 scenarios with 100 replications each. The results indicate that volatility is the most dominant factor affecting forecasting error, with significant interaction effects among structural factors. ANOVA and post hoc analysis further confirm that model performance depends more on data characteristics than on algorithmic complexity. These findings demonstrate that factorial simulation provides a systematic and robust framework for evaluating forecasting models beyond conventional empirical comparisons, while offering deeper insight into the relationship between data structure and model performance. 
Co-Authors Kusman Sadik Sri Kustiara