Chronology-Constrained Residual Option Pricing with Local Strike-Shape Diagnostics across BTC and ETH Windows
Main Article Content
Keywords
option pricing, residual learning, chronological evaluation, strike-shape diagnostics, crypto options
Abstract
Deribit option trades form an irregular sequence across strikes and expiries, so random splitting can leak later market information into model fitting. We freeze four exact-expiry 24-hour BTC and ETH windows and divide each one chronologically into 70/15/15 partitions. Eight methods predict USD prices. The deterministic comparators are fitted once; the direct multilayer perceptron, unbounded residual network, and chronology-constrained residual pricer (CCRP) are run with five seeds. SVI is fitted by exact expiry using training implied volatilities only. We also inspect prices on a fixed 15-minute reference grid for elementary bounds, monotonicity, and convexity over linear moneyness. The RMSE winner changes in every window. CCRP beats its matched unbounded residual in three of four windows, but records two wins and two losses against each of Black-Scholes, previous trade, and SVI. Its bounded map produces no bound or convexity violations on the audited grids; three monotonicity violations occur in the BTC windows. The evidence supports a limited comparison under temporal controls, not stable predictive superiority or complete arbitrage consistency for asynchronous trades.
References
- [1]F. Black and M. Scholes, “The pricing of options and corporate liabilities,” Journal of Political Economy, vol. 81, no. 3, pp. 637-654, 1973.
- [2]S. L. Heston, “A closed-form solution for options with stochastic volatility with applications to bond and currency options,” The Review of Financial Studies, vol. 6, no. 2, pp. 327-343, 1993.
- [3]J. Gatheral and A. Jacquier, “Arbitrage-free SVI volatility surfaces,” Quantitative Finance, vol. 14, no. 1, pp. 59-71, 2014.
- [4]G. Guo, A. Jacquier, C. Martini, and L. Neufcourt, “Generalized arbitrage-free SVI volatility surfaces,” SIAM Journal on Financial Mathematics, vol. 7, no. 1, pp. 619-641, 2016.
- [5]J. M. Hutchinson, A. W. Lo, and T. Poggio, “A nonparametric approach to pricing and hedging derivative securities via learning networks,” The Journal of Finance, vol. 49, no. 3, pp. 851-889, 1994.
- [6]J. De Spiegeleer, D. B. Madan, S. Reyners, and W. Schoutens, “Machine learning for quantitative finance: Fast derivative pricing, hedging and fitting,” Quantitative Finance, vol. 18, no. 10, pp. 1635-1643, 2018.
- [7]H. Buehler, L. Gonon, J. Teichmann, and B. Wood, “Deep hedging,” Quantitative Finance, vol. 19, no. 8, pp. 1271-1291, 2019.
- [8]A. Brini and J. Lenz, “Pricing cryptocurrency options with machine learning regression for handling market volatility,” Economic Modelling, vol. 136, Art. no. 106752, 2024.
- [9]L. Liang and X. Cai, “Time-sequencing European options and pricing with deep learning: Analyzing based on interpretable ALE method,” Expert Systems with Applications, vol. 187, Art. no. 115951, 2022.
- [10]T. Chen, J. Deng, and J. Nie, “Implied volatility slopes and jumps in Bitcoin options market,” Operations Research Letters, vol. 55, Art. no. 107135, 2024.
- [11]J. E. Hilliard and J. T. D. Ngo, “Bitcoin: Jumps, convenience yields, and option prices,” Quantitative Finance, vol. 22, no. 11, pp. 2079-2091, 2022.
- [12]C. Alexander and A. Imeraj, “Delta hedging Bitcoin options with a smile,” Quantitative Finance, vol. 23, no. 5, pp. 799-817, 2023.
- [13]S. N. Cohen, C. Reisinger, and S. Wang, “Arbitrage-free neural-SDE market models,” Applied Mathematical Finance, vol. 30, no. 1, pp. 1-46, 2023.
- [14]C. Bergmeir, R. J. Hyndman, and B. Koo, “A note on the validity of cross-validation for evaluating autoregressive time series prediction,” Computational Statistics & Data Analysis, vol. 120, pp. 70-83, 2018.
- [15]Deribit, “Deribit API documentation: Public market-data endpoints,” 2026. [Online]. Available: https://docs.deribit.com/. Accessed: Jul. 15, 2026.
Article Sidebar
This work is licensed under a Creative Commons Attribution 4.0 International License.