Price Dynamics and Financial Risk Analysis A Neural Hierarchical Time-Series Forecasting Approach
Keywords:
Ethereum, Deep Learning, N-HiTS, Value at Risk, PredictionAbstract
The highly volatile nature of cryptocurrency prices often causes conventional predictive models to fail in capturing complex nonlinear patterns. This study integrates the Neural Hierarchical Interpolation for Time Series Forecasting (N-HiTS) deep learning model with nonparametric Historical Simulation Value-at-Risk (VaR) method for price forecasting and risk analysis. Using univariate data on daily Ethereum closing prices from January 1, 2021, to January 31, 2025 (N = 1,491 observations), the out-of-sample evaluation was executed using a rolling cross-validation scheme initiated testing from a cut-off point in April 2024 through December 2024, where each evaluation window was set for the next 30 days. The research results show that the N-HiTS model can predict price dynamics with high accuracy, achieving an MAPE of 3.25%, an MAE of 107.825, an RMSE of 136.83, and directional accuracy of 48.28%. Risk analysis using historical simulation yielded a VaR of -6.23% at a 95% confidence level.
Downloads
References
[1] A. Afrizal, M. Marliyah, and F. Fuadi, “Analisis Terhadap Cryptocurrency (Perspektif Mata Uang, Hukum, Ekonomi Dan Syariah),” E-Mabis J. Ekon. Manaj. Dan Bisnis, vol. 22, no. 2, pp. 13–41, Nov. 2021, doi: 10.29103/e-mabis.v22i2.689.
[2] Chainalysis, “The 2024 Geography of Crypto Report,” Oct. 2024.
[3] CNN Indonesia, “OJK Catat Transaksi Kripto Tembus Rp32,45 T dari 13,71 Juta Konsumen.” [Online]. Available: https://www.cnnindonesia.com/ekonomi/20250510031904-92-1227945/ojk-catat-transaksi-kripto-tembus-rp3245-t-dari-1371-juta-konsumen
[4] finance magnates, “Crypto Young Investors: BaFin Study Reveals over 50% Trust Social Media and Finfluencers.” [Online]. Available: https://id.tradingview.com/news/financemagnates:2b834acaf094b:0-crypto-young-investors-bafin-study-reveals-over-50-trust-social-media-and-finfluencers/
[5] Rachel Narda Chaterine and Bagus Santosa, “Deretan Kasus Investasi Bodong yang Seret Nama Artis dan ‘Influencer’ Sepanjang 2022,” Kompas.com. [Online]. Available: https://nasional.kompas.com/read/2022/11/07/09105831/deretan-kasus-investasi-bodong-yang-seret-nama-artis-dan-influencer?page=all#page2
[6] R. A. Santika, Aviolla Terza Damaliana, and Mohammad Idhom, “Forecasting USD to Rupiah Exchange Rate with the Fuzzy Time Series Singh Approach,” J. Inf. Syst. Technol. Res., vol. 4, no. 3, pp. 124–134, Sep. 2025, doi: 10.55537/jistr.v4i3.1238.
[7] Aldi Bastiatul Fawait, M. Jamil, Sitti Rahmah, and Sugiarto Sugiarto, “Penerapan Metode LSTM untuk Prediksi Harga Ethereum,” vol. 9, 2025, doi: ttps://doi.org/10.30872/jurti.v9i3.22169.
[8] C. Challu, K. G. Olivares, B. N. Oreshkin, F. Garza Ramirez, M. Mergenthaler Canseco, and A. Dubrawski, “NHITS: Neural Hierarchical Interpolation for Time Series Forecasting,” Proc. AAAI Conf. Artif. Intell., vol. 37, no. 6, pp. 6989–6997, Jun. 2023, doi: 10.1609/aaai.v37i6.25854.
[9] B. N. Oreshkin, D. Carpov, N. Chapados, and Y. Bengio, “N-BEATS: Neural basis expansion analysis for interpretable time series forecasting,” Feb. 20, 2020, arXiv: arXiv:1905.10437. doi: 10.48550/arXiv.1905.10437.
[10] R. Magallanes-Quintanar, C. E. Galván-Tejada, J. I. Galván-Tejada, H. Gamboa-Rosales, S. D. J. Méndez-Gallegos, and A. García-Domínguez, “Neural Hierarchical Interpolation for Standardized Precipitation Index Forecasting,” Atmosphere, vol. 15, no. 8, p. 912, Jul. 2024, doi: 10.3390/atmos15080912.
[11] M. Saberian, V. Samadi, and I. Popescu, “Probabilistic Hierarchical Interpolation and Interpretable Configuration for Flood Prediction,” Oct. 07, 2024, Catchment hydrology/Modelling approaches. doi: 10.5194/hess-2024-261.
[12] S. F. Stefenon et al., “Neural Hierarchical Interpolation Time Series (NHITS) for Reservoir Level Multi-Horizon Forecasting in Hydroelectric Power Plants,” IEEE Access, vol. 13, pp. 54853–54865, 2025, doi: 10.1109/ACCESS.2025.3554446.
[13] K. Liao, X. Xuan, and K.-L. Ma, “Deep learning for time series forecasting: a survey of recent advances,” Front. Comput. Sci., vol. 20, no. 11, p. 2011359, Nov. 2026, doi: 10.1007/s11704-025-50947-3.
[14] X. Song, L. Deng, H. Wang, Y. Zhang, Y. He, and W. Cao, “Deep learning-based time series forecasting,” Artif. Intell. Rev., vol. 58, no. 1, p. 23, Nov. 2024, doi: 10.1007/s10462-024-10989-8.
[15] Ades Tikaningsih and Puji Lestari, “Optuna Based Hyperparameter Tuning for Improving the Performance Prediction Mortality and Hospital Length of Stay for Stroke Patients,” vol. 17, pp. 1–16, 2026, http://dx.doi.org/10.35671/telematika.v17i1.2816.
[16] A. Solihatun, La Gubu, Aswani, E. Cahyono, and L. O. Saidi, “PERHITUNGAN VALUE AT RISK (VAR) PADA PORTOFOLIO SAHAM IDX SEKTOR KEUANGAN (IDXFINANCE) MENGGUNAKAN METODE SIMULASI HISTORIS (HISTORICAL SIMULATION METHOD): PERHITUNGAN VALUE AT RISK (VAR) PADA PORTOFOLIO SAHAM IDX SEKTOR KEUANGAN,” J. Mat. Komputasi Dan Stat., vol. 3, no. 1, pp. 245–254, Apr. 2023, doi: 10.33772/jmks.v3i1.32.
[17] S. H. Al-adawiyah, E. Alisah, and A. Aziz, “Perbandingan Tingkat Akurasi Metode Average Based Fuzzy Time Series Markov Chain dan Algoritma Novel Fuzzy Time Series,” J. Ris. Mhs. Mat., vol. 1, no. 3, pp. 129–142, Feb. 2022, doi: 10.18860/jrmm.v1i3.14332.
[18] B. Wibowo and H. Fadhila, “Evaluasi Kualitas Pengungkapan Value at Risk Perbankan Indonesia,” J. Ilm. Akunt. Dan Bisnis, p. 276, May 2019, doi: 10.24843/JIAB.2019.v14.i02.p12.
[19] A. Ikhsan, T. Sutisna, and S. Widiati, “ESTIMASI RISIKO PORTOFOLIO SAHAM PERUSAHAAN PERKEBUNAN DI BURSA EFEK INDONESIA MENGGUNAKAN VALUE AT RISK NON-NORMAL,” J. Gaussian, vol. 12, no. 1, pp. 146–158, May 2023, doi: 10.14710/j.gauss.12.1.146-158.
[20] Arizal Tursina, Renea Shinta Aminda, and Immas Nurhayati, “Analisis Value at Risk (VaR) dengan Metode Historis dan Monte Carlo Dalam Harga Saham Sub Sektor Bank,” J. Ekon. Dan Bisnis, vol. 1, Oktober 2023, doi.
[21] A. Y. Febriyanti, D. A. Prasetya, and T. Trimono, “Stock Price Prediction and Risk Estimation Using Hybrid CNN-LSTM and VaR-ECF,” J. Tek. Inform. Jutif, vol. 6, no. 3, pp. 1539–1554, Jun. 2025, doi: 10.52436/1.jutif.2025.6.3.4648.
[22] B. Lim and S. Zohren, “Time-series forecasting with deep learning: a survey,” Philos. Trans. R. Soc. Math. Phys. Eng. Sci., vol. 379, no. 2194, p. 20200209, Apr. 2021, doi: 10.1098/rsta.2020.0209.
[23] F. Firmansyah, A. P. Sari, and S. Sugiarto, “Pemanfaatan Multi-Layer Perceptron (MLP) untuk Deteksi Kanker,” JASIEK J. Apl. Sains Inf. Elektron. Dan Komput., vol. 7, no. 2, pp. 127–137, Dec. 2025, doi: 10.26905/jasiek.v7i2.13438.
[24] Yahoo Finance, “Crypto Currencies: prices, changes, trading volume & daily charts.” 2024. [Online]. Available: https://finance.yahoo.com/markets/crypto/all/
[25] neuralforecast, “Nixtla.” 2021. [Online]. Available: Nixtla/neuralforecast,” GitHub, 2021. https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/nhits.py
[26] M. K. Sadadang, “ESTIMASI NILAI VALUE AT RISK PADA PORTOFOLIO SAHAM MENGGUNAKAN METODE GARCH-VINE COPULA”.
[27] I. Nabillah and I. Ranggadara, “Mean Absolute Percentage Error untuk Evaluasi Hasil Prediksi Komoditas Laut,” JOINS J. Inf. Syst., vol. 5, no. 2, pp. 250–255, Nov. 2020, doi: 10.33633/joins.v5i2.3900.
[28] H. Zhang and H. Wang, “Refitted cross-validation estimation for high-dimensional subsamples from low-dimension full data,” Sep. 21, 2024, arXiv: arXiv:2409.14032. doi: 10.48550/arXiv.2409.14032.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Vannesa Nathania, Aviolla Terza Damaliana, Shindi Shella May Wara
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
