HyperIV: Real-time Implied Volatility Smoothing

Yongxin Yang, Wenqi Chen, Chao Shu, Timothy Hospedales

TL;DR: Construct an arbitrage-free implied volatility surface from only 9 option contracts in 2 milliseconds.

Abstract: We propose HyperIV, a novel approach for real-time implied volatility smoothing that eliminates the need for traditional calibration procedures. Our method employs a hypernetwork to generate parameters for a compact neural network that constructs complete volatility surfaces within 2 milliseconds, using only 9 market observations. Moreover, the generated surfaces are guaranteed to be free of static arbitrage. Extensive experiments across 8 index options demonstrate that HyperIV achieves superior accuracy compared to existing methods while maintaining computational efficiency. The model also exhibits strong cross-asset generalization capabilities, indicating broader applicability across different market instruments. These key features -- rapid adaptation to market conditions, guaranteed absence of arbitrage, and minimal data requirements -- make HyperIV particularly valuable for real-time trading applications. We make code available at https://github.com/qmfin/hyperiv.

Lay Summary:

In the fast-paced world of options trading, a complete and accurate implied volatility surface is crucial for tasks like pricing and hedging. However, creating this surface instantly, especially using only a few current market prices (common in high-frequency trading or for illiquid assets), and ensuring it is free from theoretical inconsistencies that allow risk-free profits (arbitrage), has been a major challenge.

To address this, we developed HyperIV, an AI system that learns from vast amounts of historical market data. When presented with just a handful of live option prices – as few as nine – HyperIV instantly constructs the complete, smooth volatility surface in under two milliseconds, guaranteeing it is arbitrage-free.

HyperIV provides traders with a highly accurate and near-instantaneous view of market volatility, even with very limited real-time data. This enables faster, more reliable trading decisions and improved risk management. Its strong performance across various financial instruments also indicates broad applicability, offering a significant advancement for real-time financial modelling in today's dynamic markets.

Application-Driven Machine Learning: This submission is on Application-Driven Machine Learning.

Link To Code: https://github.com/qmfin/hyperiv

Primary Area: Applications->Everything Else

Keywords: Implied volatility surface, quantitative finance, hypernetworks

Submission Number: 271