arXiv releases arbitrage-aware probabilistic multi-step prediction paper for implied volatility surfaces
Get up to speed
Dominik Manuel Buchegger and Lukas Gonon published an updated paper on arXiv, proposing a conditional latent diffusion framework that can jointly generate 30-step trajectories of implied volatility surfaces and underlying returns. The framework learns surface representations through an arbitrage-aware autoencoder, with the diffusion process capturing evolution patterns. This framework was evaluated on the SPX surface dataset, generating realistic probabilistic multi-step scenarios, with its predicted mean outperforming the persistence prediction baseline in scenarios beyond the shortest maturity. The paper belongs to the fields of mathematical finance (q-fin.MF) and machine learning (cs.LG).
AI-generated from reports · 3 hr agoUpdated
Coverage timeline
Follow the coverage to understand different sides of the event.
- arXiv q-finArbitrage-Aware Probabilistic Multi-Step Forecasting of the Implied Volatility Surface: Jointly Generating Surface-Return Trajectories with Latent Diffusion
The study proposes a conditional latent diffusion framework that jointly generates 30-step trajectories of the implied volatility surface and underlying returns. It learns surface representations through an arbitrage-aware autoencoder, with the diffusion process capturing the evolution dynamics. The framework is evaluated on the SPX surface dataset and can generate realistic probabilistic multi-step scenarios, with its forecast mean outperforming the persistence benchmark for horizons beyond the shortest maturity.
Trend of this event
Not enough consecutive observations yet; trend is not drawn.