Overcoming Tree Extrapolation Limits in Gold Price Forecasting: A Stationary Return-Reconstructed XGBoost Approach
DOI:
https://doi.org/10.47065/jieee.v6i1.2850Keywords:
Gold Price Forecasting; XGBoost; Tree Extrapolation Limit; Stationary Return Reconstruction; Explainable AI (SHAP)Abstract
This study addresses the fundamental tree extrapolation limit inherent in gradient-boosted decision tree ensembles when forecasting non-stationary financial time series during historical market rallies. Standard raw-price regression models fail to project unseen price regimes because decision trees assign static scalar predictions to terminal leaf nodes, capping predictions at historical maximums. To overcome this spatial limitation, we propose a Stationary Return-Reconstructed XGBoost framework that shifts the machine learning target from non-stationary spot prices to stationary daily relative returns before deterministically re-integrating predicted returns into nominal price levels. Evaluated on an out-of-sample testing set spanning 2022 to 2024, the return-reconstructed framework completely eliminates flatline prediction plateaus during record-setting gold price rallies exceeding 2,800 USD per troy ounce. The proposed model achieves an R-squared score of 0.9905, a Root Mean Squared Error of 28.17 USD per troy ounce, and a Mean Absolute Percentage Error of 1.12 percent, outperforming the raw-price baseline model by 92.75 percent in average error reduction. Furthermore, game-theoretic SHAP analysis validates that daily return forecasts are driven by macroeconomically sound transmission channels, predominantly led by the US Dollar Index and monthly historical price memory
Downloads
References
Muhajir Isnin and Ira Zulfa, “Predicted Increase in Gold Price Every Year with Impact on Economic Factors,” International Journal of Economics and Management Sciences, vol. 1, no. 4, pp. 366–375, 2024, doi: 10.61132/ijems.v1i4.359.
H. Rajendran, P. Kayal, and M. Maiti, “A Multipurpose hybrid forecasting framework for economic stress scenarios: evidence from agriculture and energy sectors,” Future Business Journal, vol. 11, no. 1, pp. 1–17, 2025, doi: 10.1186/s43093-025-00612-9.
Y. Sun, Z. Qu, T. Zhang, and X. Li, “Adaptive Ensemble Learning for Financial Time-Series Forecasting: A Hypernetwork-Enhanced Reservoir Computing Framework with Multi-Scale Temporal Modeling,” Axioms, vol. 14, no. 8, pp. 1–43, 2025, doi: 10.3390/axioms14080597.
A. L. Suárez-Cetrulo, D. Quintana, and A. Cervantes, “Machine Learning for Financial Prediction Under Regime Change Using Technical Analysis: A Systematic Review,” International Journal of Interactive Multimedia and Artificial Intelligence, vol. 9, no. 1, pp. 137–148, 2024, doi: 10.9781/ijimai.2023.06.003.
S. S. Shanto and A. I. Jony, “Interpretable Ensemble Learning Approach for Predicting Student Adaptability in Online Education Environments,” Knowledge, vol. 5, no. 2, pp. 1–15, 2025, doi: 10.3390/knowledge5020010.
Z. Qi, Y. Feng, S. Wang, and C. Li, “Enhancing hydropower generation Predictions: A comprehensive study of XGBoost and Support Vector Regression models with advanced optimization techniques,” Ain Shams Engineering Journal, vol. 16, no. 1, pp. 1–16, 2025, doi: 10.1016/j.asej.2024.103206.
S. O. Abioye, Y. O. Babatunde, O. A. Abikoye, A. N. Shaibu, and B. J. Bankole, “Optimized machine learning algorithms with SHAP analysis for predicting compressive strength in high-performance concrete,” AI in Civil Engineering, vol. 4, no. 1, pp. 1–32, 2025, doi: 10.1007/s43503-025-00061-x.
W. Bao, Y. Cao, Y. Yang, H. Che, J. Huang, and S. Wen, “Data-driven stock forecasting models based on neural networks: A review,” Information Fusion, vol. 113, no. May, pp. 1–20, 2025, doi: 10.1016/j.inffus.2024.102616.
M. M. Kumbure, C. Lohrmann, P. Luukka, and J. Porras, “Machine learning techniques and data for stock market forecasting: A literature review,” Expert Syst. Appl., vol. 197, no. February, pp. 1–41, 2022, doi: 10.1016/j.eswa.2022.116659.
M. Romero Martínez, P. Carmona Ibáñez, and J. Martínez Vargas, “Predicting Business Failure with the XGBoost Algorithm: The Role of Environmental Risk,” Sustainability (Switzerland), vol. 17, no. 11, pp. 1–32, 2025, doi: 10.3390/su17114948.
M. AlGaiar, S. Bano, A. Lashin, M. Hossain, N. H. Faisal, and H. S. Abu Salem, “AiION - Novel deep learning chemical geothermometer for temperature prediction of deep geothermal reservoirs,” Renew. Energy, vol. 248, no. January, pp. 1–20, 2025, doi: 10.1016/j.renene.2025.123154.
T. T. Huynh-Cam, L. S. Chen, and H. Le, “Using decision trees and random forest algorithms to predict and determine factors contributing to first-year university students’ learning performance,” Algorithms, vol. 14, no. 11, pp. 1–17, 2021, doi: 10.3390/a14110318.
T. Blasco, J. S. Sánchez, and V. García, “A survey on uncertainty quantification in deep learning for financial time series prediction,” Neurocomputing, vol. 576, no. October, pp. 1–22, 2024, doi: 10.1016/j.neucom.2024.127339.
E. K. Ampomah, Z. Qin, and G. Nyame, “Evaluation of tree-based ensemble machine learning models in predicting stock price direction of movement,” Information (Switzerland), vol. 11, no. 6, pp. 1–21, 2020, doi: 10.3390/info11060332.
M. S. M. Bhuiyan et al., “Deep learning for algorithmic trading: A systematic review of predictive models and optimization strategies,” Array, vol. 26, no. April, pp. 1–27, 2025, doi: 10.1016/j.array.2025.100390.
P. Tziachris et al., “Spatial or Random Cross-Validation? The Effect of Resampling Methods in Predicting Groundwater Salinity with Machine Learning in Mediterranean Region,” Water (Switzerland), vol. 15, no. 12, pp. 1–18, 2023, doi: 10.3390/w15122278.
A. Nazari, M. Jamshidi, A. Roozbahani, and B. Golparvar, “Groundwater level forecasting using empirical mode decomposition and wavelet-based long short-term memory (LSTM) neural networks,” Groundw. Sustain. Dev., vol. 28, no. July, pp. 1–10, 2025, doi: 10.1016/j.gsd.2024.101397.
B. Lim, S. Ar?k, N. Loeff, and T. Pfister, “Temporal Fusion Transformers for interpretable multi-horizon time series forecasting,” Int. J. Forecast., vol. 37, no. 4, pp. 1748–1764, 2021, doi: 10.1016/j.ijforecast.2021.03.012.
M. Zaher, A. S. Ghoneim, L. Abdelhamid, and A. Atia, “Unlocking the potential of RNN and CNN models for accurate rehabilitation exercise classification on multi-datasets,” Multimed. Tools Appl., vol. 84, no. 3, pp. 1261–1301, 2025, doi: 10.1007/s11042-024-19092-0.
V. Hassija et al., “Interpreting Black-Box Models: A Review on Explainable Artificial Intelligence,” Cognit. Comput., vol. 16, no. 1, pp. 45–74, 2024, doi: 10.1007/s12559-023-10179-8.
Y. Zhou, Z. Ma, and L. Fu, “A Review of Key Signal Processing Techniques for Structural Health Monitoring: Highlighting Non-Parametric Time-Frequency Analysis, Adaptive Decomposition, and Deconvolution,” Algorithms, vol. 18, no. 6, pp. 1–48, 2025, doi: 10.3390/a18060318.
G. Grekousis, “Geographical-XGBoost: a new ensemble model for spatially local regression based on gradient-boosted trees,” J. Geogr. Syst., vol. 27, no. 2, pp. 169–195, 2025, doi: 10.1007/s10109-025-00465-4.
F. Petropoulos et al., “Operational Research: methods and applications,” Journal of the Operational Research Society, vol. 75, no. 3, pp. 423–617, 2024, doi: 10.1080/01605682.2023.2253852.
S. S Band et al., “Application of explainable artificial intelligence in medical health: A systematic review of interpretability methods,” Inform. Med. Unlocked, vol. 40, no. March, pp. 1–20, 2023, doi: 10.1016/j.imu.2023.101286.
L. Ruan and H. Jiang, “Stock Price Prediction Using FinBERT-Enhanced Sentiment with SHAP Explainability and Differential Privacy,” Mathematics, vol. 13, no. 17, pp. 1–22, 2025, doi: 10.3390/math13172747.
Bila bermanfaat silahkan share artikel ini
Berikan Komentar Anda terhadap artikel Overcoming Tree Extrapolation Limits in Gold Price Forecasting: A Stationary Return-Reconstructed XGBoost Approach
ARTICLE HISTORY
Issue
Section
Copyright (c) 2026 Imam Saputra, Zulfikar, Juni Ismail, Alfry Aristo Jansen Sinlae, Yanto Saputra, Raja Anan Nasution, Elsy Rahajeng, Mesran
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under Creative Commons Attribution 4.0 International License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (Refer to The Effect of Open Access).
