Evaluasi Prediksi Harga Saham Nokia Menggunakan LSTM Univariat dengan Pendekatan Walk-Forward Validation
Keywords:
Stock Price Prediction, Univariate LSTM, Walk-Forward Validation, Model Performance, MulticollinearityAbstract
Prediksi harga saham merupakan tantangan sentral di pasar modal yang kompleks dan volatil.
Meskipun model pembelajaran mendalam seperti Long Short-Term Memory (LSTM) telah
menunjukkan potensi, banyak penelitian mengabaikan masalah multikolinearitas pada model
multivariat dan menggunakan metode evaluasi yang tidak realistis. Untuk mengatasi ini,
penelitian ini mengembangkan model prediksi harga saham Nokia menggunakan arsitektur
LSTM univariat yang hanya memanfaatkan harga penutupan, sebuah keputusan yang
didasarkan pada bukti empiris multikolinearitas tinggi antar fitur harga. Kinerja model
dievaluasi secara ketat menggunakan Walk-Forward Validation (WFV) untuk mensimulasikan
kondisi perdagangan nyata dan menghindari bias evaluasi. Hasilnya menunjukkan performa
yang sangat baik dan stabil, dengan model mampu menjelaskan 94.46% varians data (R² =
0.9446) dan mencapai Mean Absolute Percentage Error (MAPE) sebesar 2.75%. Konsistensi
ini terbukti melalui 30 iterasi WFV, yang mengonfirmasi ketahanan model di berbagai kondisi
pasar. Penelitian ini menyimpulkan bahwa pendekatan model LSTM univariat yang dievaluasi
dengan WFV terbukti efektif dan andal, bahkan dapat menjadi pilihan superior dibandingkan
model yang lebih kompleks. Temuan ini menegaskan bahwa relevansi fitur dan standar
evaluasi yang ketat lebih krusial daripada kompleksitas arsitektur, memberikan kontribusi
metodologis penting bagi pengembangan model prediksi finansial yang andal di masa depan.
References
Aldhyani, T. H. H., & Alzahrani, A. (2022). Framework for Predicting and Modeling Stock Market Prices Based on Deep Learning Algorithms. Electronics (Switzerland), 11(19). https://doi.org/10.3390/electronics11193149
Ali, M., Khan, D. M., Alshanbari, H. M., & El-Bagoury, A. A. A. H. (2023). Prediction of Complex Stock Market Data Using an Improved Hybrid EMD-LSTM Model. Applied Sciences (Switzerland), 13(3). https://doi.org/10.3390/app13031429
Ammer, M. A., & Aldhyani, T. H. H. (2022). Deep Learning Algorithm to Predict Cryptocurrency Fluctuation Prices: Increasing Investment Awareness. Electronics (Switzerland), 11(15). https://doi.org/10.3390/electronics11152349
Bielskis, A., & Belovas, I. (2022). Comparative analysis of stock price ARIMA and LSTM forecasting methods. Lietuvos Matematikos Rinkinys, 63. https://doi.org/10.15388/lmr.2022.29755
Chandola, D., Mehta, A., Singh, S., Tikkiwal, V. A., & Agrawal, H. (2023). Forecasting Directional Movement of Stock Prices using Deep Learning. Annals of Data Science, 10(5). https://doi.org/10.1007/s40745-022-00432-6
Cheng, C.-H., Tsai, M.-C., & Chang, C. (2022). A Time Series Model Based on Deep Learning and Integrated Indicator Selection Method for Forecasting Stock Prices and Evaluating Trading Profits. Systems, 10(6), 243. https://doi.org/10.3390/systems10060243
Freeborough, W., & van Zyl, T. (2022). Investigating Explainability Methods in Recurrent Neural Network Architectures for Financial Time Series Data. Applied Sciences, 12(3), 1427. https://doi.org/10.3390/app12031427
Jarrah, M., & Derbali, M. (2023). Predicting Saudi Stock Market Index by Using Multivariate Time Series Based on Deep Learning. Applied Sciences (Switzerland), 13(14). https://doi.org/10.3390/app13148356
Kim, J., Kim, H.-S., & Choi, S.-Y. (2023). Forecasting the S&P 500 Index Using Mathematical-Based Sentiment Analysis and Deep Learning Models: A FinBERT Transformer Model and LSTM. Axioms, 12(9), 835. https://doi.org/10.3390/axioms12090835
Lawi, A., Mesra, H., & Amir, S. (2022). Implementation of Long Short-Term Memory and Gated Recurrent Units on grouped time-series data to predict stock prices accurately. Journal of Big Data, 9(1), 89. https://doi.org/10.1186/s40537-022-00597-0
Low, P. R., & Sakk, E. (2023). Comparison between autoregressive integrated moving average and long short term memory models for stock price prediction. IAES International Journal of Artificial Intelligence, 12(4), 1828–1835. https://doi.org/10.11591/ijai.v12.i4.pp1828-1835
Pang, T. (2024). APPL stock price prediction based on LSTM and GRU. Applied and Computational Engineering, 47(1), 200–206. https://doi.org/10.54254/2755-2721/47/20241343
Peng, Z.-Y., & Guo, P.-C. (2022). A Data Organization Method for LSTM and Transformer When Predicting Chinese Banking Stock Prices. Discrete Dynamics in Nature and Society, 2022(1). https://doi.org/10.1155/2022/7119678
Ulum, D. S. N., & Girsang, A. S. (2022). Hyperparameter Optimization of Long-Short Term Memory using Symbiotic Organism Search for Stock Prediction. International Journal of Innovative Research and Scientific Studies, 5(2), 121–133. https://doi.org/10.53894/ijirss.v5i2.415
Vásquez Sáenz, J., Quiroga, F. M., & Bariviera, A. F. (2023). Data vs. information: Using clustering techniques to enhance stock returns forecasting. International Review of Financial Analysis, 88, 102657. https://doi.org/10.1016/j.irfa.2023.102657
Zhang, C., Sjarif, N. N. A., & Ibrahim, R. (2024). 1D-CapsNet-LSTM: A deep learning-based model for multi-step stock index forecasting. Journal of King Saud University - Computer and Information Sciences, 36(2), 101959. https://doi.org/10.1016/j.jksuci.2024.101959
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 JURNAL TEKNIK INFORMATIKA UNIS
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
