English

Arbitrage-Free Implied Volatility Surface Generation with Variational Autoencoders

Mathematical Finance 2022-01-31 v3 Machine Learning Computational Finance Pricing of Securities Machine Learning

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.

Cite

@article{arxiv.2108.04941,
  title  = {Arbitrage-Free Implied Volatility Surface Generation with Variational Autoencoders},
  author = {Brian Ning and Sebastian Jaimungal and Xiaorong Zhang and Maxime Bergeron},
  journal= {arXiv preprint arXiv:2108.04941},
  year   = {2022}
}

Comments

20 pages, 7 figures

R2 v1 2026-06-24T05:00:30.416Z