Target-Characteristic-Aware Forecasting of Fuel–Equity Rolling Correlations: Evidence That PCA and Persistence Outperform Hybrid Deep Learning Models

Authors

  • Chrisnawan Prastya Atmaja Universitas Amikom Yogyakarta
  • Ferian Fauzi Abdulloh

DOI:

10.33395/sinkron.v10i3.16410

Keywords:

Attention-LSTM, CNN-LSTM, Financial Forecasting, Fuel-Equity Correlation, PCA, Persistence Model, Rolling Correlation

Abstract

Background: Forecasting dynamic fuel–equity market correlations is important for financial forecasting because fuel price movements can affect market risk, investor sentiment, and cross-market stability. However, most previous studies focus on direct price or volatility prediction, while the forecasting of rolling correlations between multiple fuel types and global equity indices remains less explored. Objective: This study analyzes and predicts time-varying correlations between global fuel prices and major stock market indices by comparing baseline, PCA-based, standalone deep learning, and hybrid deep learning models. Methods: The proposed framework applies data preprocessing, return transformation, 60-day rolling correlation construction across three fuel variables (petrol, diesel, LPG) and five stock indices (S&P 500, NASDAQ, FTSE 100, Nikkei 225, IHSG/JKSE), normalization, Principal Component Analysis (PCA), sequence generation, and comparative evaluation of 15 forecasting models. Model performance was measured using RMSE, MAE, R-squared (R²), and Directional Accuracy. Results: The PCA Model achieved the lowest RMSE of 0.016909 and the highest R² of 0.967045, while the Persistence Model produced the lowest MAE of 0.002293 and the highest Directional Accuracy of 97.767280%. Hybrid deep learning models showed higher errors and negative R² values, indicating weaker performance on smooth and persistent rolling correlation targets. Conclusion: The findings show that architectural complexity does not necessarily improve forecasting performance when the target series is smooth and persistent. This study contributes empirical evidence and a replicable comparative framework showing that model selection in financial time-series forecasting should consider target characteristics, particularly smoothness and temporal persistence, rather than relying solely on complex deep learning architectures.

GS Cited Analysis

Downloads

Download data is not yet available.

References

Akusta, A., Yıldırım, H. H., Gün, M., & Sakarya, Ş. (2026). Deep learning enhanced energy market prediction: A robust ARIMAX–LSTM fusion for crude oil pricing. Journal of Computational and Applied Mathematics, 474, 117006. https://doi.org/10.1016/j.cam.2025.117006

Amagbo, R., & Geman, H. (2026). Dynamic and asymmetric spillovers between crude oil, biofuels and agricultural commodities: Evidence from periods of geopolitical tensions and energy policy uncertainty. International Review of Economics & Finance, 108, 105276. https://doi.org/10.1016/j.iref.2026.105276

Beck, N., Schemm, J., Ehrig, C., Sonnleitner, B., Neumann, U., & Zimmermann, H. G. (2025). Mind the naive forecast! A rigorous evaluation of forecasting models for time series with low predictability. Applied Intelligence. https://doi.org/10.1007/s10489-025-06268-w

Ben Flah, I., Samet, K., El Ammari, A., & Terzi, C. (2025). Shock and Volatility Transmissions Across Global Commodity and Stock Markets Spillovers: Empirical Evidence from Africa. Journal of Risk and Financial Management, 18(6), 332. https://doi.org/10.3390/jrfm18060332

Ben Romdhane, W., & Boubaker, H. (2026). A Hybrid HAR-LSTM-GARCH Model for Forecasting Volatility in Energy Markets. Journal of Risk and Financial Management,19(1),77.https://doi.org/10.3390/jrfm19010077

Cao, L., et al. (2016). Forecasting daily stock market return using dimensionality reduction. Expert Systems with Applications, 67, 126–139. https://doi.org/10.1016/j.eswa.2016.09.027

Chen, Z., et al. (2024). PCA-ICA-LSTM: A hybrid deep learning model based on dimension reduction methods to predict S&P 500 index price. Computational Economics. https://doi.org/10.1007/s10614-024-10629-x

Choudhary, J., Sharma, H. K., Malik, P., & Majumder, S. (2025). Price Forecasting of Crude Oil Using Hybrid Machine Learning Models. Journal of Risk and Financial Management, 18(7),346.https://doi.org/10.3390/jrfm18070346

Chung, V., Espinoza, J., & Quispe, R. (2025). Forecasting Financial Volatility Under Structural Breaks: A Comparative Study of GARCH Models and Deep Learning Techniques. Journal of Risk and Financial Management, 18(9), 494. https://doi.org/10.3390/jrfm18090494

Dang, T. P. T., & Liu, B. (2026). The spillover effects of gold and oil prices: a study of the Vietnamese and global markets. Journal of Financial Economic Policy, 1–18. https://doi.org/10.1108/JFEP-07-2025-0291

Egbe, A. A., Eyupoglu, S. Z., & Seraj, M. (2026). Sustainability Investment in Distress: Volatility Spillovers and ESG Markets Portfolio Implications. Sustainability, 18(9), 4403. https://doi.org/10.3390/su18094403

Firouzjaee, M. A., et al. (2024). The interpretability of LSTM models for predicting oil company stocks: Impact of correlated features. International Journal of Energy Research. https://doi.org/10.1155/2024/5526692

Kakade, K., Jain, I., & Mishra, A. K. (2022). Value-at-Risk forecasting: A hybrid ensemble learning GARCH-LSTM based approach. Resources Policy,78,102903.https://doi.org/10.1016/j.resourpol.2022.102903

Lahmiri, S. (2026). A hybrid CNN-LSTM-GRU system tuned by whale optimization algorithm for enhanced gold and brent price prediction. Finance Research Open, 2(1), 100100.https://doi.org/10.1016/j.finr.2026.100100

Li, A., & Zhong, B. (2025). Asymmetric spillover connectedness between clean energy markets and industrial stock markets: How uncertainties affect it. PLoS ONE, 20(3), e0316171. https://doi.org/10.1371/journal.pone.0316171

Li, S., & Xu, S. (2025). Enhancing stock price prediction using GANs and transformer-based attention mechanisms. Empirical Economics, 68, 373–403. https://doi.org/10.1007/s00181-024-02644-6

Liu, Z., & Wang, J. (2026). Stock markets, Sino–US monetary policy, and international crude oil prices. Finance Research Letters, 94, 109629. https://doi.org/10.1016/j.frl.2026.109629

Ma, Q., & Han, M. (2026). Forecasting Dynamic Correlations Between Carbon, Energy, and Stock Markets Using a BOHB-Optimized Multivariable Graph Neural Network. Mathematics,14(1),171.https://doi.org/10.3390/math14010171

Mishra, A., & Dash, A. K. (2024). Return volatility of Asian stock exchanges; a GARCH DCC analysis with reference of Bitcoin and global crude oil price movement. Journal of Chinese Economic and Foreign Trade Studies, 17(1), 29–48. https://doi.org/10.1108/JCEFTS-01-2024-0009

Nguyen, A. T. H., Nguyen, P. T. T., Pham, U. T. T., Le, A. T., Le, A. T. N., Tran, T. P., & Nguyen, O. V. H. (2026). Dynamic volatility spillovers during market turbulence: examining the interconnectedness among gold, oil, cryptocurrencies and stock markets. Journal of International Economics and Management, 1–16. https://doi.org/10.1108/JIEM-04-2025-0003

Ni, J., & Xu, Y. (2023). Forecasting the Dynamic Correlation of Stock Indices Based on Deep Learning Method. Computational Economics, 61(1), 35–55. https://doi.org/10.1007/s10614-021-10198-3

Oyewola, D. O., Akinwunmi, S. A., & Omotehinwa, T. O. (2024). Deep LSTM and LSTM-Attention Q-learning based reinforcement learning in oil and gas sector prediction. Knowledge-Based Systems, 284, 111290. https://doi.org/10.1016/j.knosys.2023.111290

Rahmi, A., & Haryanto, T. (2025). Examining the link between world oil price volatility and sectoral stock performance in Indonesia. Cogent Economics & Finance, 13(1). https://doi.org/10.1080/23322039.2025.2535484

Sen, A., & Dutta Choudhury, K. (2024). Forecasting the Crude Oil prices for last four decades using deep learning approach. Resources Policy, 88, 104438. https://doi.org/10.1016/j.resourpol.2023.104438

Siami-Namini, S., Tavakoli, N., & Namin, A. S. (2019). The performance of LSTM and BiLSTM in forecasting time series. In Proceedings of the 2019 IEEE International Conference on Big Data (pp. 3285–3292). IEEE. https://doi.org/10.1109/BigData47090.2019.9005997

Su, Y., & Feng, X. (2026). Machine learning approaches to predicting energy price correlation: From a responsible AI perspective. Technological Forecasting and Social Change,225,124515.https://doi.org/10.1016/j.techfore.2025.124515

Wang, X. (2024). Index prediction model based on LASSO-PCA and deep learning. International Journal of Crowd Science. https://doi.org/10.26599/IJCS.2024.9100013

Yang, A. (2025). Big data-driven corporate financial forecasting and decision support: A study of CNN-LSTM machine learning models. Frontiers in Applied Mathematics and Statistics, 11. https://doi.org/10.3389/fams.2025.1566078

Zhang, X., et al. (2024). CNN-CBAM-LSTM: Enhancing stock return prediction through long and short information mining. Mathematics, 12(23), 3738. https://doi.org/10.3390/math12233738

Zheng, L., Sun, Y., & Wang, S. (2024). A novel interval-based hybrid framework for crude oil price forecasting and trading. Energy Economics, 130, 107266. https://doi.org/10.1016/j.eneco.2023.107266

Downloads


Crossmark Updates

How to Cite

Atmaja, C. P., & Ferian Fauzi Abdulloh. (2026). Target-Characteristic-Aware Forecasting of Fuel–Equity Rolling Correlations: Evidence That PCA and Persistence Outperform Hybrid Deep Learning Models. Sinkron : Jurnal Dan Penelitian Teknik Informatika, 10(3), 1984-1996. https://doi.org/10.33395/sinkron.v10i3.16410

Issue

Vol. 10 No. 3 (2026): Article Research July 2026

Section

Articles

License

Copyright (c) 2026 Chrisnawan Prastya Atmaja, Ferian Fauzi Abdulloh

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Crossref
Scopus
Google Scholar
Europe PMC