Arbitrage-Free Implied Volatility Surface Generation with Variational Autoencoders

DOI10.1137/21M1443546MaRDI QIDQ6070669zbMATH OpenOpenAlexFDO

Authors Brian Ning, Sebastian Jaimungal, Xiaorong Zhang, Maxime Bergeron

Publication date 23 November 2023

Published in SIAM Journal on Financial Mathematics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2108.04941

zbMATH Keywords

computational financeimplied volatilityLévy processesmachine learningvariational autoencoder

Mathematics Subject Classification ID

Processes with independent increments; Lévy processes (60G51) Applications of stochastic analysis (to PDEs, etc.) (60H30) Artificial neural networks and deep learning (68T07) Derivative securities (option pricing, hedging, etc.) (91G20)

Abstract: We propose a hybrid method for generating arbitrage-free implied volatility (IV) surfaces consistent with historical data by combining model-free Variational Autoencoders (VAEs) with continuous time stochastic differential equation (SDE) driven models. We focus on two classes of SDE models: regime switching models and L'evy additive processes. By projecting historical surfaces onto the space of SDE model parameters, we obtain a distribution on the parameter subspace faithful to the data on which we then train a VAE. Arbitrage-free IV surfaces are then generated by sampling from the posterior distribution on the latent space, decoding to obtain SDE model parameters, and finally mapping those parameters to IV surfaces. We further refine the VAE model by including conditional features and demonstrate its superior generative out-of-sample performance.

Recommendations

• FuNVol: multi-asset implied volatility market simulator using functional principal components and neural SDEs
• Simulation of Arbitrage-Free Implied Volatility Surfaces
• A two-step framework for arbitrage-free prediction of the implied volatility surface
• Deep weighted Monte Carlo: a hybrid option pricing framework using neural networks
• Asymptotics of implied volatility to arbitrary order

Cites work

• A closed-form solution for options with stochastic volatility with applications to bond and currency options
• Calculation of the Wasserstein Distance Between Probability Distributions on the Line
• Deep learning volatility: a deep neural network perspective on pricing and calibration in (rough) volatility models
• Fourier space time-stepping for option pricing with Lévy models
• Hedging Option Books Using Neural-SDE Market Models
• scientific article; zbMATH DE number 1742902 (Why is no real title available?)
• scientific article; zbMATH DE number 2015362 (Why is no real title available?)
• Lévy Processes and Stochastic Calculus
• Nonparametric kernel density estimation and its computational aspects
• Optimal Transport
• OPTION PRICING FOR TRUNCATED LÉVY PROCESSES
• Short communication: Beyond surrogate modeling: learning the local volatility via shape constraints
• Tangent Lévy market models

Cited in

(5)

• Simulation of Arbitrage-Free Implied Volatility Surfaces
• FuNVol: multi-asset implied volatility market simulator using functional principal components and neural SDEs
• VolGAN: A Generative Model for Arbitrage-Free Implied Volatility Surfaces
• Deep volatility imputation using convolutional autoencoders
• Joint arbitrage-free smoothing of American call and put options surfaces

This page was built for publication: Arbitrage-Free Implied Volatility Surface Generation with Variational Autoencoders

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6070669)