Comparison results for stochastic volatility models via coupling (Q2430255)

The motivation for the authors comes from the fact that the famous Black-Scholes model, or the geometric Brownian motion, is universally acknowledged, but fails to capture many observed features of financial data. For instance, the volatility of stock prices changes with time and more importantly from the derivative pricing perspective, the market does not price consistently in the Black-Scholes model. The authors consider in their paper stochastic volatility models. Their aim is to investigate the properties of such models, and to find out to what extent and with regard to which models the properties of the geometric Brownian motion carry over to a stochastic volatility setting. They are particularly eager to answer questions of the form: can the process hit zero; does the discounted price process converge; are discounted prices true martingales; are option prices convex in the underlying; are the option prices monotonic in the model parameters? The main results of this paper are a construction of the solution to a stochastic volatility model, the application of this construction to derive results describing when the discounted asset price can hit zero, and when it is a martingale, and a comparison theorem for option prices in different stochastic volatility models. These theoretical results are augmented by the discussion of several examples, including both well-known models form the literature and the new models which illustrate the various phenomena.