Conformal Prediction Intervals for Daily Equity Returns Under Volatility Regimes
DOI:
https://doi.org/10.24002/jbi.v17i2.15190Keywords:
conformal prediction, prediction intervals, equity returns, volatility regimes, coverage backtestingAbstract
Risk reports rely on interval forecasts, yet parametric volatility models lose their nominal coverage when markets turn turbulent. We ask whether distribution-free conformal prediction can give dependable one-step-ahead intervals for daily equity returns across volatility regimes. Under one online protocol we compare an unconditional Gaussian benchmark, Gaussian and Student-t GARCH(1,1) intervals, a volatility-scaled split conformal predictor, and adaptive conformal inference, alongside five machine learning interval methods, on four European indices and the S&P 500 from 1990 to 2026. The unconditional benchmark overcovers yet covers only about 85 percent of high-volatility days. The GARCH intervals are efficient and well calibrated; the conformal methods give the most accurate marginal coverage and the best coverage on turbulent days. Significance tests and a sensitivity analysis confirm that volatility-scaled conformal correction is most reliable in turbulence, at a small cost in width.
Laporan risiko bergantung pada prakiraan interval, tetapi model volatilitas parametrik kehilangan cakupan nominalnya ketika pasar bergejolak. Kami menguji apakah prediksi konformal yang bebas distribusi dapat memberikan interval satu langkah ke depan yang andal untuk imbal hasil saham harian di berbagai rezim volatilitas. Di bawah satu protokol daring, kami membandingkan tolok ukur Gaussian tak bersyarat, interval GARCH(1,1) Gaussian dan Student-t, prediktor konformal terbelah berskala volatilitas, dan inferensi konformal adaptif, bersama lima metode interval berbasis pembelajaran mesin, pada empat indeks Eropa dan indeks S&P 500 dari 1990 hingga 2026. Tolok ukur tak bersyarat memiliki cakupan berlebih secara keseluruhan, namun hanya mencakup sekitar 85 persen hari bervolatilitas tinggi. Interval GARCH efisien dan terkalibrasi baik, sedangkan metode konformal memberikan kalibrasi marginal paling akurat dan cakupan terbaik pada hari bergejolak. Uji signifikansi dan analisis sensitivitas menegaskan bahwa koreksi konformal berskala volatilitas paling andal saat bergejolak, dengan biaya lebar yang kecil.
References
[1] N. Cakici, C. Fieberg, D. Metko, and A. Zaremba, "Predicting returns with machine learning across horizons, firm size, and time," Journal of Financial Data Science, vol. 5, no. 4, pp. 119–144, 2023, doi: 10.3905/jfds.2023.1.139.
[2] W. Drobetz and T. Otto, "Empirical asset pricing via machine learning: evidence from the European stock market," Journal of Asset Management, vol. 22, no. 7, pp. 507–538, 2021, doi: 10.1057/s41260-021-00237-x.
[3] A. Marsi, "Predicting European stock returns using machine learning," SN Business & Economics, vol. 3, no. 7, 2023, doi: 10.1007/s43546-023-00487-4.
[4] J. Shi and H. Yin, "Machine learning in multifactor modeling: a critical assessment," Journal of Beta Investment Strategies, vol. 17, no. 1, pp. 7–19, 2026, doi: 10.3905/jbis.2026.1.105.
[5] T. Liu, "A comparative study of transformer-based and classical models for financial time-series forecasting," Journal of Risk and Financial Management, vol. 19, no. 3, 2026, doi: 10.3390/jrfm19030203.
[6] T. Cipra, Time Series in Economics and Finance. Cham, Switzerland: Springer, 2020, doi: 10.1007/978-3-030-46347-2.
[7] T. Kim, J. Kim, Y. Tae, C. Park, J. H. Choi, and J. Choo, "Reversible instance normalization for accurate time-series forecasting against distribution shift," in Proceedings of the 10th International Conference on Learning Representations (ICLR), 2022.
[8] S. Maitra, "Adaptive fractional differencing: theory and methodology," IEEE Access, vol. 13, pp. 209,312–209,327, 2025, doi: 10.1109/ACCESS.2025.3633728.
[9] S. Maitra, "Adaptive fractional differencing for learning-optimal long-memory preservation," IEEE Access, vol. 14, pp. 32,407–32,430, 2026, doi: 10.1109/ACCESS.2026.3666622.
[10] J. Wan, N. Xia, Y. Yin, X. Pan, J. Hu, and J. Yi, "TCDformer: a transformer framework for non-stationary time series forecasting based on trend and change-point detection," Neural Networks, vol. 173, 2024, doi: 10.1016/j.neunet.2024.106196.
[11] H. Fischer, Á. Blanco-Fernández, and P. Winker, "Predicting stock return volatility: can we benefit from regression models for return intervals?," Journal of Forecasting, vol. 35, no. 2, pp. 113–146, 2016, doi: 10.1002/for.2371.
[12] L. Hu and Y. Shen, "A predictive analytics approach for forecasting global stock index returns using deep learning techniques," Decision Analytics Journal, vol. 18, 2026, doi: 10.1016/j.dajour.2026.100685.
[13] B. Candelon, G. Colletaz, C. Hurlin, and S. Tokpavi, "Backtesting value-at-risk: a GMM duration-based test," Journal of Financial Econometrics, vol. 9, no. 2, pp. 314–343, 2011, doi: 10.1093/jjfinec/nbq025.
[14] K. Retzlaff, F. Schlembach, D. Bams, and P. Dreesen, "Testing marginal and conditional coverage in conformal prediction for non-stationary time series via value-at-risk backtesting," Proceedings of Machine Learning Research, vol. 266, pp. 725–747, 2025.
[15] T. Kaiser and P. Herzog, "A tutorial on distribution-free uncertainty quantification using conformal prediction," Advances in Methods and Practices in Psychological Science, vol. 8, no. 4, 2025, doi: 10.1177/25152459251380452.
[16] N. Dewolf, B. De Baets, and W. Waegeman, "Valid prediction intervals for regression problems," Artificial Intelligence Review, vol. 56, no. 1, pp. 577–613, 2023, doi: 10.1007/s10462-022-10178-5.
[17] U. Johansson, H. Boström, T. Löfström, and H. Linusson, "Regression conformal prediction with random forests," Machine Learning, vol. 97, no. 1–2, pp. 155–176, 2014, doi: 10.1007/s10994-014-5453-0.
[18] V. Vovk, I. Petej, I. Nouretdinov, V. Manokhin, and A. Gammerman, "Computationally efficient versions of conformal predictive distributions," Neurocomputing, vol. 397, pp. 292–308, 2020, doi: 10.1016/j.neucom.2019.10.110.
[19] I. I. Akpabio and S. A. Savari, "On uses of noise analysis for the uncertainty quantification of line edge roughness estimation," IEEE Transactions on Semiconductor Manufacturing, vol. 36, no. 3, pp. 319–326, 2023, doi: 10.1109/TSM.2023.3270230.
[20] M. Fontana, G. Zeni, and S. Vantini, "Conformal prediction: a unified review of theory and new challenges," Bernoulli, vol. 29, no. 1, pp. 1–23, 2023, doi: 10.3150/21-BEJ1447.
[21] P. Hajibabaee, F. Pourkamali-Anaraki, and M. A. Hariri-Ardebili, "Adaptive conformal prediction intervals using data-dependent weights with application to seismic response prediction," IEEE Access, vol. 12, pp. 53,579–53,597, 2024, doi: 10.1109/ACCESS.2024.3387858.
[22] I. Gibbs and E. Candès, "Conformal inference for online prediction with arbitrary distribution shifts," Journal of Machine Learning Research, vol. 25, 2024.
[23] M. Zaffran, O. Féron, Y. Goude, J. Josse, and A. Dieuleveut, "Adaptive conformal predictions for time series," Proceedings of Machine Learning Research, vol. 162, pp. 25,834–25,866, 2022.
[24] A. Makhin and M. Panov, "Automatic adaptive conformal inference for time series forecasting," Lecture Notes in Computer Science, vol. 15419, pp. 262–274, 2025, doi: 10.1007/978-3-031-88036-0_16.
[25] A. Farinhas, C. Zerva, D. Ulmer, and A. F. T. Martins, "Non-exchangeable conformal risk control," in Proceedings of the 12th International Conference on Learning Representations (ICLR), 2024.
[26] R. F. Barber and A. Pananjady, "Predictive inference for time series: why is split conformal effective despite temporal dependence?," Proceedings of Machine Learning Research, vol. 313, 2026.
[27] P. L. Seabe, C. R. B. Moutsinga, and M. Aphane, "Uncertainty-first forecasting of the South African equity market using deep learning and temporal conformal prediction," Big Data and Cognitive Computing, vol. 10, no. 3, 2026, doi: 10.3390/bdcc10030093.
[28] X. Zhou, B. Chen, Y. Gui, and L. Cheng, "Conformal prediction: a data perspective," ACM Computing Surveys, vol. 57, no. 7, 2025, doi: 10.1145/3714380.
[29] L. de Azevedo Takara, R. C. T. de Souza, V. C. Mariani, L. dos S. Coelho, and S. F. Stefenon, "Multi-scale convolutional Chebyshev Kolmogorov-Arnold networks using conformal prediction for photovoltaic power forecasting," IEEE Access, vol. 14, pp. 81,170–81,194, 2026, doi: 10.1109/ACCESS.2026.3694780.
[30] S. Ghimire, T. Nguyen-Huy, R. C. Deo, D. Casillas-Pérez, R. Prasad, and S. Salcedo-Sanz, "Point and confidence interval prediction of electricity price based on hybrid dual-decomposition and deep learning model," Energy, vol. 350, 2026, doi: 10.1016/j.energy.2026.140692.
[31] W. Sun, F. Tack, L. Clarisse, C. Clerbaux, and M. Van Roozendael, "Inferring surface NO₂ over Western Europe: a machine learning approach with uncertainty quantification," Journal of Geophysical Research: Atmospheres, vol. 129, no. 4, 2024, doi: 10.1029/2023JD039820.
[32] L. Li, H. Wang, Y. Liu, and F. Zhang, "Probabilistic interval prediction method based on shape-adaptive quantile regression," Expert Systems, vol. 41, no. 8, 2024, doi: 10.1111/exsy.13585.
[33] P. Kumar, P. Das, and E. Ghaderpour, "Decomposed and hybridized deep learning models for an application to stock price prediction," in Proceedings of the International Conference on Modeling, Simulation & Intelligent Computing (MoSICom), 2023, pp. 609–613, doi: 10.1109/MoSICom59118.2023.10458816.
[34] Z. Zhang, Q. Liu, Y. Hu, and H. Liu, "Multi-feature stock price prediction by LSTM networks based on VMD and TMFG," Journal of Big Data, vol. 12, no. 1, 2025, doi: 10.1186/s40537-025-01127-4.
[35] S. Kumar, A. Rao, and M. Dhochak, "Hybrid ML models for volatility prediction in financial risk management," International Review of Economics & Finance, vol. 98, 2025, doi: 10.1016/j.iref.2025.103915.
[36] A. S. Al-Jawarneh, A. R. M. Alsayed, and H. N. Ayyoub, "Optimizing modelling accuracy using variational mode decomposition and elastic net regression: evidence in stock market prediction," Array, vol. 28, 2025, doi: 10.1016/j.array.2025.100603.
[37] Y. Wang, X. Hao, and C. Wu, "Forecasting stock returns: a time-dependent weighted least squares approach," Journal of Financial Markets, vol. 53, 2021, doi: 10.1016/j.finmar.2020.100568.
[38] Y. Huang and C. Yang, "Enhancing stock price forecasting with a modular deep learning framework incorporating plug-and-play transformer variants," Expert Systems with Applications, vol. 315, 2026, doi: 10.1016/j.eswa.2026.131572.
[39] R. A. Andika and C. Dewi, "Importance of feature selection for multiple disease classification," Jurnal Buana Informatika, vol. 16, no. 1, pp. 34–45, 2025, doi: 10.24002/jbi.v16i01.11354.
[40] S. Alawiyah, D. Yuliati, and N. Ulinnuha, "Implementasi LightGBM dengan KNN imputation untuk deteksi dini risiko kehamilan," Jurnal Buana Informatika, vol. 17, no. 1, pp. 21–32, 2026, doi: 10.24002/jbi.v17i1.14454.
[41] M. Buczyński and M. Chlebus, "Old-fashioned parametric models are still the best: a comparison of value-at-risk approaches in several volatility states," Journal of Risk Model Validation, vol. 14, no. 2, pp. 1–20, 2020, doi: 10.21314/JRMV.2020.222.
[42] Z. Du, P. Pei, X. Wang, and T. Yang, "Powerful backtests for historical simulation expected shortfall models," Journal of Business & Economic Statistics, vol. 42, no. 3, pp. 864–874, 2024, doi: 10.1080/07350015.2023.2252881.
[43] T. Bollerslev, “Generalized autoregressive conditional heteroskedasticity,” Journal of Econometrics, vol. 31, no. 3, pp. 307–327, 1986, doi: 10.1016/0304-4076(86)90063-1.
[44] P. R. Hansen and A. Lunde, “A forecast comparison of volatility models: does anything beat a GARCH(1,1)?,” Journal of Applied Econometrics, vol. 20, no. 7, pp. 873–889, 2005, doi: 10.1002/jae.800.
[45] A. Wilhelmsson, “GARCH forecasting performance under different distribution assumptions,” Journal of Forecasting, vol. 25, no. 8, pp. 561–578, 2006, doi: 10.1002/for.1009.
[46] U. Johansson, H. Boström, and T. Löfström, "Investigating normalized conformal regressors," in Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI), 2021, doi: 10.1109/SSCI50451.2021.9659853.
[47] H. Arian, D. Norouzi Mobarekeh, and L. Seco, "Backtest overfitting in the machine learning era: a comparison of out-of-sample testing methods in a synthetic controlled environment," Knowledge-Based Systems, vol. 305, 2024, doi: 10.1016/j.knosys.2024.112477.
[48] C. Spulbǎr and C. C. Ene, "Rolling walk-forward LSTM for daily stock-return prediction in European defence markets," Studies in Business and Economics, vol. 20, no. 3, pp. 240–260, 2025, doi: 10.2478/sbe-2025-0053.
[49] A. Arasteh, "A data-driven prediction method for multi-period portfolio optimization using the real options approach," Finance Research Letters, vol. 80, 2025, doi: 10.1016/j.frl.2025.107403.
[50] I. Nouretdinov, "Conformal test martingales for hypergraphical models," Annals of Mathematics and Artificial Intelligence, 2024, doi: 10.1007/s10472-024-09951-9.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Copyright of this journal is assigned to Jurnal Buana Informatika as the journal publisher by the knowledge of author, whilst the moral right of the publication belongs to author. Every printed and electronic publications are open access for educational purposes, research, and library. The editorial board is not responsible for copyright violation to the other than them aims mentioned before. The reproduction of any part of this journal (printed or online) will be allowed only with a written permission from Jurnal Buana Informatika.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
