Multi-Asset Option Pricing Using Normal Tempered Stable Processes with Stochastic Correlation

We present a new model of normal tempered stable (NTS) processes with stochastic correlation for multi-asset option pricing. The model is constructed by extending the constant correlation term in the NTS model to the stochastic correlation making use of the Ornstein-Uhlenbeck process. As the proposed model captures the stylized empirical facts of heavy tails, kurtosis, skewness, and stochastic correlation, multi-asset option pricing is one of the various applications. In particular, we discuss its application to quanto option pricing as a bivariate example and derive its closed-form solution under the risk-neutral measure. Building on the calibration results, we identify the presence of stochastic correlation in the risk-neutral world