MOMENT EXPLOSIONS AND LONG-TERM BEHAVIOR OF AFFINE STOCHASTIC VOLATILITY MODELS

Martin Keller‐Ressel

Mathematical Finance · 2010 · 137 citations · 24 references

Concepts

TL;DR

The paper studies asset‑pricing models where the log‑price and stochastic variance form an affine process under risk‑neutral dynamics. The authors derive conditions for conservativeness and martingale property of the price process and give explicit formulas for the explosion times of its moments. They obtain the invariant distribution of the variance, analyze moment‑explosion implications for implied‑volatility asymptotics, and illustrate the results on the Heston, Bates, and Barndorff‑Nielsen–Shephard models.

Abstract

We consider a class of asset pricing models, where the risk-neutral joint process of log-price and its stochastic variance is an affine process in the sense of Duffie, Filipovic, and Schachermayer. First we obtain conditions for the price process to be conservative and a martingale. Then we present some results on the long-term behavior of the model, including an expression for the invariant distribution of the stochastic variance process. We study moment explosions of the price process, and provide explicit expressions for the time at which a moment of given order becomes infinite. We discuss applications of these results, in particular to the asymptotics of the implied volatility smile, and conclude with some calculations for the Heston model, a model of Bates and the Barndorff-Nielsen–Shephard model.

References

24