Volatility Surface Reconstruction using Deep Learning under No-Arbitrage Constraints
Researchers have developed deep learning models to reconstruct implied volatility surfaces from limited and noisy option data, adhering to no-arbitrage constraints. The study compared various neural network architectures, including Transformers and U-Nets, against traditional SVI parameterizations using real market data. Findings indicate that Transformer and U-Net models offer superior reconstruction accuracy, especially when data is sparse, and that incorporating soft arbitrage penalties effectively reduces violations with only a minor impact on accuracy. AI
IMPACT This research could lead to more accurate financial modeling and risk assessment in quantitative finance.