Overview
Stochastic Volatility Models model the volatility of a process as a stochastic process in its own right. Because it is known that the volatility of commodity is not static, adding randomness to the volatility can make the model more realistic.
Eydeland and Geman
Eydeland and Geman propose the following model.
{% dS(t) = \mu S(t) + \sqrt{V(t)} S(t) dW_1(t) %}
{% dV(t) = a(\bar{V} - V(t)) dt + \eta \sqrt{V(t)}dW_2(t) %}
see Farelli chpt 2