arXiv q-fin· Aleksandar Arandjelovic, Patrick Cheridito·· 1 days agoAI Score41
On Continuity and Layering of Conditional Distributions for Derivative Pricing
On the continuity of conditional distributions and stratification for derivative pricing
AI summary
This arXiv quantitative finance paper focuses on constructing neural network layering factors for derivative pricing under continuous-time models. It establishes continuity results for conditional expectations and conditional distributions, provides counterexamples to continuity theory, and derives a universal approximation theorem for admissible layering factors based on neural network density results. Numerical experiments show that the learned layering factors outperform all baselines in two derivative pricing scenarios.
Source:arXiv q-fin · arxiv.org