BBCA Stock Price Forecasting Using Statistical, Machine Learning, and Deep Learning Models with Technical Indicator Features: A Comparative Analysis

Rendi Prasetya; Herlinda Herlinda

doi:10.30998/faktor.exac.v19i1.2036

Authors

Rendi Prasetya Author
Herlinda Herlinda Author

DOI:

https://doi.org/10.30998/faktor.exac.v19i1.2036

Keywords:

Prediksi harga saham, Machine learning, Deep learning, Random forest, LSTM

Abstract

Due to high volatility, non-linearity, and complex dynamics, stock price forecasting remains a challenging task in financial time-series analysis. This study quantitatively compares the forecasting performance of six methods: Moving Average (MA), Linear Regression (LR), Support Vector Regression (SVR), Random Forest (RF), Gaussian Process Regression (GPR), and Long Short-Term Memory (LSTM), utilizing daily BBCA stock data from May 2011 to April 2026. Preprocessing encompassed data validation, chronological ordering, Min-Max normalization, and feature engineering via lagged price values, MA, and the Relative Strength Index (RSI). Model evaluation relied on Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE), supplemented by historical visualization and a 30-trading-day future projection. Experimental results indicate that GPR achieved superior performance (MAE: 58.68, MSE: 8,940.29, RMSE: 94.55, MAPE: 1.12%). RF yielded competitive results (MAE: 113.36, MSE: 86,609.19, RMSE: 294.29, MAPE: 1.54%), followed by LSTM (MAE: 88.75, MSE: 16,269.79, RMSE: 127.55, MAPE: 1.92%). Conversely, LR and SVR exhibited higher prediction errors, failing to capture complex price movements. Visualizations confirm that GPR, RF, and LSTM closely mirrored actual price trends, with GPR and LSTM generating more stable 30-day trajectories than linear or conventional statistical approaches. Overall, non-linear learning-based models, particularly GPR, offer reliable alternatives for short-term stock forecasting.

Downloads

Download data is not yet available.

References

[1] D. Mishra, P. Sinha, dan R. Singh, “Stock Price Prediction using Machine Learning and Deep Learning Techniques: A Review,” J. Phys. Conf. Ser., vol. 1950, no. 1, hlm. 12019, 2021.

[2] M. R. Shah dan S. Mehmood, “Stock Price Prediction Using Machine Learning Algorithms: A Comprehensive Survey,” Journal of Business and Economics Review, vol. 8, no. 1, hlm. 1–10, 2023.

[3] B. Purwanto dan H. Fitriyani, “Perbandingan Metode Simple Moving Average dan Linear Regression dalam Prediksi Harga Saham,” Jurnal Manajemen Keuangan, vol. 7, no. 1, hlm. 1–10, 2018.

[4] R. Wiyono dan S. Hartati, “Analisis Perbandingan Metode Linear Regression dan Moving Average untuk Prediksi Harga Saham,” Jurnal Penelitian Komunikasi, vol. 20, no. 2, hlm. 163–176, 2017.

[5] S. N. Handayani dan D. Kurniawan, “Prediksi Harga Saham Menggunakan Support Vector Regression dengan Optimasi Parameter Grid Search,” Jurnal Teknologi Informasi dan Ilmu Komputer, vol. 8, no. 1, hlm. 123–130, 2021.

[6] S. A. Ghafor dan A. B. Abdulbaqi, “Stock Price Prediction Using Machine Learning Models: A Comparative Study,” International Journal of Advanced Computer Science and Applications, vol. 14, no. 3, hlm. 45–53, 2023.

[7] D. E. Nugroho dan D. R. Putra, “Implementasi Support Vector Regression (SVR) untuk Prediksi Harga Saham IHSG,” Jurnal Rekayasa Teknologi Informasi, vol. 3, no. 2, hlm. 143–150, 2019.

[8] A. Almahfud dan D. Sumarsono, “Implementasi Long Short-Term Memory (LSTM) untuk Prediksi Harga Saham,” Jurnal Ilmiah Komputer dan Informatika, vol. 9, no. 2, hlm. 123–131, 2020.

[9] R. Setiawan dan E. Winarko, “Prediksi Harga Saham Menggunakan LSTM dan Metode Feature Engineering,” Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi), vol. 4, no. 2, hlm. 346–352, 2020.

[10] Z. Yan dan J. Meng, “Stock Price Prediction Based on LSTM Neural Network,” J. Phys. Conf. Ser., vol. 1732, no. 1, hlm. 12028, 2021.

[11] M. K. Azhar dan E. Winarko, “Prediksi Harga Saham Menggunakan Long Short-Term Memory (LSTM) dengan Optimasi Adam,” Jurnal Informatika Ekonomi Bisnis, vol. 1, no. 2, hlm. 115–125, 2022.

[12] R. H. Fariz dan T. B. Adji, “Kombinasi Metode ARIMA dan Long Short-Term Memory (LSTM) untuk Prediksi Harga Saham,” Jurnal Rekayasa Sistem Industri, vol. 10, no. 1, hlm. 1–12, 2021.

[13] J. Chen, Z. Zhang, J. Cao, dan Y. Li, “Stock Price Prediction Based on Hybrid Attention Mechanism and Bi-directional LSTM,” Journal of Finance and Economics, vol. 5, no. 1, hlm. 1–10, 2020.

[14] M. Li, X. Huang, dan S. Zhu, “A Hybrid Deep Learning Model for Stock Price Prediction Based on CNN-BiLSTM and Attention Mechanism,” Expert Syst. Appl., vol. 216, hlm. 119330, 2023.

[15] X. Zhu dan M. Li, “Multi-task Learning for Stock Price Prediction Based on CNN-BiLSTM and Attention Mechanism,” Expert Syst. Appl., vol. 235, hlm. 121115, 2024.

[16] M. Hussain, A. Ali, dan S. Hussain, “Stock Market Prediction Using Machine Learning: A Survey,” Journal of Financial Studies, vol. 1, no. 1, hlm. 1–15, 2018.

[17] A. Hasibuan dan I. M. Sari, “Perbandingan Algoritma Random Forest dan Linear Regression dalam Prediksi Harga Saham,” Jurnal Ilmu Komputer dan Sistem Informasi, vol. 5, no. 1, hlm. 1–8, 2022.

[18] R. Saputra, “Prediksi Harga Saham Menggunakan Gaussian Process Regression,” Jurnal Sistem dan Teknologi Informasi, vol. 8, no. 1, hlm. 1–8, 2019.

[19] W. Khan dan A. Muhammad, “Stock Market Prediction Using LSTM Deep Learning Model,” International Journal of Computer Science and Network Security, vol. 20, no. 4, hlm. 168–175, 2020.

[20] B. Xu, J. Yu, dan B. Fu, “Stock Price Forecasting Model Based on CEEMDAN-CNN-BiLSTM-Attention Mechanism,” J. Phys. Conf. Ser., vol. 2198, no. 1, hlm. 12020, 2022.

[21] J. Liu, Y. Li, dan Y. Chen, “Stock Price Prediction Based on BiLSTM and Technical Indicators,” IEEE Access, vol. 10, hlm. 10329–10338, 2022.

[22] X. Li, W. Jiang, J. Chen, X. Li, dan C. Liu, “A Survey on Blockchain Consensus Algorithms,” Concurr. Comput., vol. 32, no. 15, hlm. e5788, 2020, doi: 10.1002/cpe.5788.