Neural Networks Compute 0DTE Option Prices And Greeks
On March 8, 2026, Takayuki Sakuma posts an arXiv preprint presenting a differential machine learning method for zero-days-to-expiry (0DTE) options under a stochastic-volatility jump-diffusion model. The method computes prices and Greeks in a single network evaluation using a Black–Scholes representation with maturity-gated variance correction, a PIDE-residual penalty, and a separate jump-operator network trained in three stages. Bates-model simulations report improved jump-term approximation, enhanced Greeks accuracy, stable one-day delta hedges, and faster runtimes versus a Fourier-based benchmark.
Key Points
- 1Introduce neural differential ML method computing 0DTE option prices and Greeks in one network evaluation
- 2Represent prices via Black–Scholes form with maturity-gated variance correction and PIDE-residual penalty
- 3Provide jump-operator network and three-stage training, improving Greeks accuracy and enabling stable hedges
Scoring Rationale
Strong methodological novelty and practical speedups, limited by being a single arXiv preprint without peer review.
Sources
Public references used for this report.
Practice with real Ride-Hailing data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Ride-Hailing problems
