Regime-Sensitive VaR Forecasting for Indonesian Gold with Quantile LSTM: Walk-Forward Backtesting and Stress Diagnostics
DOI:
https://doi.org/10.52436/1.jutif.2026.7.5.5740Keywords:
Indonesian gold, Quantile LSTM, Stress diagnostics, Value-at-Risk, Walk-forward backtestingAbstract
This study investigates regime-sensitive Value-at-Risk (VaR) forecasting for ANTAM Indonesian gold using a Quantile LSTM under a rigorous walk-forward evaluation protocol. Daily prices are transformed into log-returns, and volatility- and momentum-based features are constructed in a time-respecting manner to prevent information leakage. The model is trained with multi-quantile pinball loss to estimate one-day-ahead VaR at and . Out-of-sample performance is evaluated using a walk-forward expanding-window scheme with periodic retraining, and stress-regime diagnostics are conducted via rolling-volatility-based regime labeling to assess stability during turbulent episodes. Quantile LSTM is benchmarked against standard VaR baselines (Historical Simulation, parametric normal, EWMA, and GARCH(1,1) with normal and Student-t innovations) using violation rates, Kupiec and Christoffersen backtests, pinball loss, and loss-based model comparison. Results indicate good aggregate calibration for Quantile LSTM, with violation rates of 0.554% (6/1083) for VaR 1% and 5.079% (55/1083) for VaR 5%, passing unconditional coverage. However, at there are indications of clustered exceedances, highlighting the value of rolling and regime-based diagnostics. At the 1% tail, GARCH-t achieves lower and significantly different pinball losses, reinforcing heavy-tail models as strong benchmarks for extreme risk. Beyond accuracy metrics, the regime-sensitive diagnostics provide actionable signals for deploying Quantile LSTM within automated risk-monitoring pipelines, where rolling exceedance alarms and adaptive retraining can improve operational reliability under volatility regime shifts.
Downloads
References
M. Zaenal Asikin, “Peran Emas sebagai Lindung Nilai terhadap Ketidakpastian Pasar Keuangan Global,” Hawalah Kaji. Ilmu Ekon. Syariah, vol. 3, no. 3, pp. 123–133, 2024, doi: 10.57096/hawalah.v3i3.54.
M. A. Madani and Z. Ftiti, “Is gold a hedge or safe haven against oil and currency market movements? A revisit using multifractal approach,” Ann. Oper. Res., vol. 313, no. 1, pp. 367–400, 2022, doi: 10.1007/s10479-021-04288-6.
Y. Zhu, S. I. Taasim, and A. Daud, “Volatility Modeling and Tail Risk Estimation of Financial Assets: Evidence from Gold, Oil, Bitcoin, and Stocks for Selected Markets,” Risks, vol. 13, no. 7, pp. 1–30, 2025, doi: 10.3390/risks13070138.
Z. Kamalia, W. T. Utami, and A. Susilo, “Pemodelan Prediksi Harga Saham Emas ANTAM Menggunakan Gated Recurrent Unit dan Regresi Linear Berganda pada Time Series,” TECHNOMEDIA Informatics Comput. Sci., vol. 3, no. 1, pp. 8–14, 2026, doi: 10.58641.
M. Zainuddin and dan Amrul Mutaqin, “Konvergensi Finansial Syariah: Sinergi Emas, Inklusi Digital, dan Ketahanan Sosial-Ekonomi dalam Praktik Pegadaian Syariah (Islamic Financial Convergence: Gold Synergies, Digital Inclusion, and Socio-Economic Resilience in Islamic Pawnshop Practices),” J. Islam. Account. Financ., vol. 5, no. 1, pp. 1–20, 2025, doi: 10.30162/al-muhasib.v5i1.2233.
A. Z. Kamalia, C. N. Latansa, and Zaenur Rozikin, “Klasifikasi Kondisi Pasar Harga Emas ANTAM Indonesia Menggunakan Algoritma Decision Tree,” JUKTISI, vol. 4, no. 3, pp. 2087–2098, 2026, doi: 10.62712/juktisi.v3i3.307.
D. Mikindani, J. O’Brien, P. Leahy, and P. Deeney, “The financial risks from wind turbine failures: a value at risk approach,” Appl. Econ., vol. 57, no. 39, pp. 6105–6120, 2025, doi: 10.1080/00036846.2024.2380542.
G. Keilbar and W. Wang, “Modelling systemic risk using neural network quantile regression,” Empir. Econ., vol. 62, no. 1, pp. 93–118, 2022, doi: 10.1007/s00181-021-02035-1.
S. D. Subramoney, K. Chinhamu, and R. Chifurira, “Value at Risk long memory volatility models with heavy-tailed distributions for cryptocurrencies,” Front. Appl. Math. Stat., vol. 11, pp. 1–17, 2025, doi: 10.3389/fams.2025.1567626.
S. Dlamini and S. C. Shongwe, “Comparative Analysis of Tail Risk in Emerging and Developed Equity Markets : An Extreme Value Theory Perspective,” Financ. Stud., vol. 14, no. 11, pp. 1–34, 2026, doi: 10.3390/ijfs14010011.
G. Chakraborty, G. R. Chandrashekhar, and G. Balasubramanian, “Measurement of extreme market risk: Insights from a comprehensive literature review,” Cogent Econ. Financ., vol. 9, no. 1, pp. 1–24, 2021, doi: 10.1080/23322039.2021.1920150.
E. Samunderu and Y. T. Murahwa, “Return Based Risk Measures for Non-Normally Distributed Returns: An Alternative Modelling Approach,” J. Risk Financ. Manag., vol. 14, no. 11, pp. 2007–2008, 2021, doi: 10.3390/jrfm14110540.
M. C. Lee, “A Hybrid EGARCH–Informer Model with Consistent Risk Calibration for Volatility and CVaR Forecasting,” Mathematics, vol. 13, no. 19, pp. 1–22, 2025, doi: 10.3390/math13193108.
M. Rostami, F. Rasti, and E. Abbasi, “Copula-Based Risk Modeling: A Comparative Analysis of MCAViaR and Gaussian Copulas for Global Indices,” J. Math. Model. Financ., vol. 5, no. 2, pp. 77–106, 2025, doi: 10.22054/jmmf.2025.86227.1187.
M. Hajiabbasi Somehsaraie, S. Tofighi, Z. Wang, J. Wang, and S. Xiao, “A New Loss Function for Enhancing Peak Prediction in Time Series Data with High Variability,” Forecasting, vol. 7, no. 4, pp. 1–21, 2025, doi: 10.3390/forecast7040075.
Z. Cui, T. Li, Z. Ding, X. Li, and J. Wu, “Probabilistic oil price forecasting with a variational mode decomposition-gated recurrent unit model incorporating pinball loss,” Data Sci. Manag., vol. 8, no. 3, pp. 237–247, 2025, doi: 10.1016/j.dsm.2024.10.003.
A. G. A. Savada, G. F. Nama, T. Yulianti, and M. Mardiana, “Peramalan Data Ekonomi Menggunakan Model Hybrid Vector Autoregressive-Long Short Term Memory,” J. Tek. Inform. dan Sist. Inf., vol. 11, no. 1, pp. 91–104, 2025, doi: 10.28932/jutisi.v11i1.10066.
K. Astudillo, M. Flores, M. Soliz, G. Ferreira, and J. Varela-Aldás, “A Hybrid GAS-ATT-LSTM Architecture for Predicting Non-Stationary Financial Time Series,” Mathematics, vol. 13, no. 14, pp. 1–29, 2025, doi: 10.3390/math13142300.
M. Lopez‐martin, A. Sanchez‐esguevillas, L. Hernandez‐callejo, J. I. Arribas, and B. Carro, “Additive ensemble neural network with constrained weighted quantile loss for probabilistic electric‐load forecasting,” Sensors, vol. 21, no. 9, pp. 1–37, 2021, doi: 10.3390/s21092979.
M. S. A. Dolon, “Hybrid Machine Learning–Driven Financial Forecasting Models: Integrating Lstm, Prophet, and Xgboost for Enhanced Stock Price and Risk Prediction,” Rev. Appl. Sci. Technol., vol. 04, no. 01, pp. 01–34, 2025, doi: 10.63125/nr1j8527.
G. Fatouros, G. Makridis, D. Kotios, J. Soldatos, M. Filippakis, and D. Kyriazis, “DeepVaR: a framework for portfolio risk assessment leveraging probabilistic deep neural networks,” Digit. Financ., vol. 5, no. 1, pp. 29–56, 2023, doi: 10.1007/s42521-022-00050-0.
K. Olorunnimbe and H. Viktor, Deep learning in the stock market—a systematic survey of practice, backtesting, and applications, vol. 56, no. 3. Springer Netherlands, 2023. doi: 10.1007/s10462-022-10226-0.
L. Padmakumari and M. Shaik, “An Empirical Investigation of Value at Risk (VaR) Forecasting Based on Range-Based Conditional Volatility Models,” Eng. Econ., vol. 34, no. 3, pp. 275–292, 2023, doi: 10.5755/j01.ee.34.3.30335.
R. Kwong and H. Wong, “Value-at-risk in the presence of asset price bubbles,” J. Appl. Econ., vol. 25, no. 1, pp. 361–384, 2022, doi: 10.1080/15140326.2021.1927441.
F. Roccazzella, P. Gambetti, and F. Vrins, “Optimal and robust combination of forecasts via constrained optimization and shrinkage,” Int. J. Forecast., vol. 38, no. 1, pp. 97–116, 2022, doi: 10.1016/j.ijforecast.2021.04.002.
F. Bagheri, D. R. Recupero, and E. Sirnes, “Leveraging Return Prediction Approaches for Improved Value-at-Risk Estimation,” Data, vol. 8, no. 8, 2023, doi: 10.3390/data8080133.
H. Wang and H. Ma, “Optimal Investment Portfolios for Internet Money Funds Based on LSTM and La-VaR: Evidence from China,” Mathematics, vol. 10, no. 16, pp. 1–18, 2022, doi: 10.3390/math10162864.
A. Aslam, R. K. Brahmana, and X. W. Yeap, “The Feature Importance of Quantile Connectedness among Islamic Sectoral Indices and Meme Assets,” Asia-Pacific Financ. Mark., vol. 1, pp. 1–5, 2025, doi: 10.1007/s10690-025-09558-y.
J. Wang and Z. Chen, “Exploring Low-Risk Anomalies: A Dynamic CAPM Utilizing a Machine Learning Approach,” Mathematics, vol. 11, no. 14, pp. 1–22, 2023, doi: 10.3390/math11143220.
M. Wooh and P. Cho, “Integrating Vision Transformer and Time–Frequency Analysis for Stock Volatility Prediction,” Mathematics, vol. 13, no. 23, p. 3787, 2025, doi: 10.3390/math13233787.
M. Sahiner, Forecasting volatility in Asian financial markets: evidence from recursive and rolling window methods, vol. 2, no. 10. Springer International Publishing, 2022. doi: 10.1007/s43546-022-00329-9.
C. Wongoutong, “The impact of neglecting feature scaling in k-means clustering,” PLoS One, vol. 19, no. 12, pp. 1–19, 2024, doi: 10.1371/journal.pone.0310839.
Triyono, L. Soviyanti, Sutrisno, and M. F. Kamarudin, “Income Risk Analysis and Sustainability of Shallot Farming in Bantul Yogyakarta , Indonesia,” vol. 01005, pp. 1–19, 2026, doi: 10.1051/bioconf/202620801005.
Z. Hendri, S. Shuib, A. Widarjono, and F. Sintarini, “Cash Flow Risk and Financial Balance: Evidence from Islamic Rural Banks in Indonesia,” Econ. J. Ekon. Islam, vol. 16, no. 1, pp. 171–190, 2025, doi: 10.21580/economica.2025.16.1.25835.
N. Azliana Aridi, T. Siow Hooi, and C. Wen Cheong, “A comparative VaR analysis between low-frequency and high-frequency conditional EVT models during COVID-19 crisis,” Cogent Econ. Financ., vol. 12, no. 1, pp. 1–19, 2024, doi: 10.1080/23322039.2024.2377495.
A. S. Baig, H. A. Butt, and R. Khalid, “Estimating value-at-risk models for non-conventional equity market index,” Rev. Financ. Econ., vol. 40, no. 1, pp. 63–76, 2022, doi: 10.1002/rfe.1139.
K. K. Lawuobahsumo, B. Algieri, and A. Leccadito, “Forecasting cryptocurrencies returns: Do macroeconomic and financial variables improve tail expectation predictions?,” Qual. Quant., vol. 58, no. 3, pp. 2647–2675, 2024, doi: 10.1007/s11135-023-01761-1.
D. Fantazzini, “Adaptive Conformal Inference for Computing Market Risk Measures: An Analysis with Four Thousand Crypto-Assets,” J. Risk Financ. Manag., vol. 17, no. 248, pp. 1–44, 2024, doi: 10.3390/jrfm17060248.
Additional Files
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Antika Zahrotul Kamalia, Arief Wibowo, Rusdah

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

</a



