Stochastic volatility models recognize that the volatility of an asset price is not constant but rather changes randomly over time. This approach contrasts with simpler models like Black-Scholes, which assume a fixed volatility parameter. Stochastic volatility models are designed to capture real-world market phenomena such as volatility clustering and mean reversion.
Model
A stochastic volatility model typically involves two coupled stochastic processes: one for the underlying asset price and another for its instantaneous variance. The Heston model is a prominent example, where the variance process follows a mean-reverting square-root process. These models provide a more accurate representation of asset price dynamics, especially for long-term options and complex derivatives.
Dynamics
The dynamics of stochastic volatility models allow for the generation of volatility smiles and skews, which are commonly observed in options markets. By incorporating a correlation parameter between the asset price and its volatility, these models can explain why implied volatility tends to increase when the asset price decreases. This dynamic behavior is crucial for accurate risk management and pricing of exotic options.
