Estimating risks of option books using neural-SDE market models

In this paper, we examine the capacity of an arbitrage-free neural-SDE market model to produce realistic scenarios for joint dynamics multiple European options on a single underlying. We subsequently demonstrate its use as risk simulation engine option portfolios. Through backtesting analysis, show that our models are more computationally efficient and accurate evaluating Value-at-Risk (VaR) portfolios, with better coverage performance less procyclicality than standard filtered historical approaches.