Basket Options with Volatility Skew: Calibrating a Local Volatility Model by Sample Rearrangement

TL;DR

This paper introduces a fast, sample-rearrangement-based method to calibrate local volatility models for basket options, effectively capturing complex asset dependencies and enabling accurate, rapid pricing of basket derivatives.

Contribution

It presents an efficient sample rearrangement technique to calibrate multivariate local volatility models for baskets, improving speed and accuracy in derivative pricing.

Findings

Calibration achieved within seconds

Near-perfect fit to market index options

Effective modeling of asset dependency structures

Abstract

The pricing of derivatives tied to baskets of assets demands a sophisticated framework that aligns with the available market information to capture the intricate non-linear dependency structure among the assets. We describe the dynamics of the multivariate process of constituents with a copula model and propose an efficient method to extract the dependency structure from the market. The proposed method generates coherent sets of samples of the constituents process through systematic sampling rearrangement. These samples are then utilized to calibrate a local volatility model (LVM) of the basket process, which is used to price basket derivatives. We show that the method is capable of efficiently pricing basket options based on a large number of basket constituents, accomplishing the calibration process within a matter of seconds, and achieving near-perfect calibration to the index options of the market.