MODEL HIBRIDA SARIMA–GRU UNTUK PERAMALAN HARGA SAHAM PT TELKOM INDONESIA TBK
Rika Qodriah(1*); Martanto Martanto(2); Raditya Danar Dana(3); Dadang Sudrajat(4); Saeful Anwar(5);
(1) STMIK IIMI
(2) STMIK IIMI
(3) STMIK IIMI
(4) STMIK IIMI
(5) STMIK IIMI
(*) Corresponding Author
AbstractPenelitian ini dilatarbelakangi oleh tantangan dalam memprediksi harga saham pada pasar negara berkembang, termasuk saham PT Telkom Indonesia Tbk (TLKM), yang memiliki karakteristik volatil, nonstasioner, serta menunjukkan kombinasi pola musiman linier dan dinamika residual nonlinier. Tujuan penelitian ini adalah mengembangkan dan mengevaluasi model prediksi berbasis pendekatan hibrida Seasonal Autoregressive Integrated Moving Average – Gated Recurrent Unit (SARIMA–GRU) yang dirancang untuk menangkap struktur linier-musiman sekaligus ketergantungan nonlinier pada data deret waktu. Metode penelitian menggunakan pendekatan kuantitatif dengan desain eksperimen terstruktur, meliputi tahap pengumpulan data melalui Yahoo Finance API, pra-pemrosesan, pemodelan SARIMA, pelatihan model GRU pada residual, serta integrasi prediksi hibrida. Hipotesis penelitian menyatakan bahwa model SARIMA–GRU mampu menghasilkan kesalahan prediksi yang lebih rendah dibandingkan model SARIMA atau GRU secara individual. Hasil evaluasi menggunakan RMSE dan MAE menunjukkan bahwa model hibrida memberikan peningkatan akurasi dengan RMSE lebih rendah dibanding model linier tunggal, menandakan bahwa kombinasi kedua pendekatan lebih adaptif dalam menangani dinamika pasar yang kompleks. Secara keseluruhan, penelitian ini membuktikan bahwa model SARIMA–GRU merupakan pendekatan yang efektif untuk peramalan harga saham TLKM dan berpotensi diterapkan dalam sistem pendukung keputusan investas Keywordsperamalan harga saham; SARIMA-GRU; model hibrida; deret waktu; deep learning
|
Full Text:PDF |
Article MetricsAbstract view: 154 timesPDF view: 97 times |
Digital Object Identifier https://doi.org/10.33096/linier.v2i4.3345
|
Cite |
References
M. Karanasos, S. Yfanti, and J. Hunter, “Emerging stock market volatility and economic fundamentals: the importance of US uncertainty spillovers, financial and health crises,” Ann Oper Res, vol. 313, no. 1, pp. 25–49, 2022, doi: 10.1007/s10479-021-04042-y.
T. Gudmundsson and K. Sethi, “Emerging markets: prospects and challenges,” Global Business Review, 2022, doi: 10.1177/09718907221100345.
M. Saberironaghi, J. Ren, and A. Saberironaghi, “Stock market prediction using machine learning and deep learning techniques: A review,” AppliedMath, vol. 5, no. 3, p. 76, 2025, doi: 10.3390/appliedmath5030076.
W. Chen, “Deep learning for financial time series prediction: A survey,” Knowledge Based Systems, vol. 263, p. 110111, 2023, doi: 10.1016/j.knosys.2023.110111.
K. Aggarwal and M. K. Jha, “Stock returns seasonality in emerging Asian markets,” Asia Pacific Financial Markets, vol. 30, no. 1, pp. 109–130, 2023, doi: 10.1007/s1069002209370y.
Y. Si, “Modeling opening price spread of Shanghai Composite Index based on ARIMA GRU/LSTM hybrid model [Working paper],” SSRN, 2023. doi: 10.2139/ssrn.4493799.
D. S. O. Júnior, M. Ferreira, and M. Fernandes, “A hybrid system based on ensemble learning to model time series data,” Inf Sci (N Y), vol. 634, pp. 1–18, 2023, doi: 10.1016/j.ins.2023.104184.
D. Stempień and R. Ślepaczuk, “Hybrid models for financial forecasting: Combining econometric, machine learning, and deep learning models,” Financial Innovation, 2025, doi: 10.1007/s41060-025-00854-4.
Z. Dong and Y. Zhou, “A novel hybrid model for financial forecasting based on CEEMDAN-SE and ARIMA-CNN-LSTM,” Mathematics, vol. 12, no. 16, p. 2434, 2024, doi: 10.3390/math12162434.
S. S. W. Fatima, “A review of time series forecasting algorithms for industrial systems,” Machines, vol. 12, no. 6, p. 380, 2024, doi: 10.3390/2075-1702/12/6/380.
L. C. de Jesus Jr., F. Fernández Navarro, and M. Carbonero Ruz, “Enhancing financial time series forecasting through topological data analysis,” Neural Comput Appl, vol. 37, pp. 6527–6545, 2025, doi: 10.1007/s00521-024-10787-x.
S. Meisenbacher, “Review of automated time series forecasting pipelines,” WIREs Data Mining and Knowledge Discovery, vol. 12, no. 6, p. e1475, 2022, doi: 10.1002/widm.1475.
W. Long, J. Gao, K. Bai, and Z. Lu, “A hybrid model for stock price prediction based on multi view heterogeneous data,” Financial Innovation, vol. 10, no. 48, 2024, doi: 10.1186/s40854-023-00519-w.
M. I. Tabash and et al., “Market Shocks and Stock Volatility: Evidence from Developed and Emerging Markets,” Risks, vol. 12, no. 1, p. 2, 2024, doi: 10.3390/risks12010002.
M. J. Ayala, N. Gonzálvez Gallego, and R. Arteaga Sánchez, “Google search volume index and investor attention in stock market: a systematic review,” Financial Innovation, vol. 10, no. 70, 2024, doi: 10.1186/s40854-023-00606-y.
P. Balasubramanian and et al., “A systematic literature survey on recent trends in stock market forecasting,” Applied Sciences, vol. 14, no. 8, p. 5183, 2024, doi: 10.3390/app14085183.
M. Kreuzer, L. Hoffmann, T. Schneider, and R. Berger, “Seasonal-trend decomposition and nonlinear residual analysis for financial time-series forecasting,” Journal of Computational Finance and Data Analytics, vol. 14, no. 2, pp. 112–128, 2025.
M. L. Hossain, S. M. N. Shams, and S. M. Ullah, “Time series and deep learning approaches for renewable energy forecasting in Dhaka: A comparative study of ARIMA, SARIMA, and LSTM models,” Sustainability, vol. 6, p. 775, 2025, doi: 10.1007/s43621025017335.
S. C. Necula and et al., “Advanced hybrid models for air pollution forecasting: Combining SARIMA, BiLSTM and attention mechanisms,” Electronics (Basel), vol. 14, no. 3, p. 549, 2025, doi: 10.3390/electronics14030549.
A. Mahmoud and et al., “Leveraging Hybrid Deep Learning Models for Enhanced Time Series Forecasting,” Neural Comput Appl, 2024, doi: 10.1007/s11063-024-11656-3.
A. Lama and et al., “Python code for modelling ARIMA LSTM architecture with random forest technique,” Inf Sci (N Y), 2024, doi: 10.1016/j.ins.2024.04.038.
Refbacks
- There are currently no refbacks.
Copyright (c) 2026 Rika Qodriah, Martanto Martanto, Raditya Danar Dana, Dadang Sudrajat, Saeful Anwar
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Linier: Literatur Informatika dan Komputer indexed by
ISSN 3063-2218
Published by Prodi Teknik Informatika FIK Universitas Muslim Indonesia
Website : https://jurnal.fikom.umi.ac.id/index.php/LINIER/
Email : linier@umi.ac.id
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0