Stochastic integration with respect to Volterra processes

We construct the basis of a stochastic calculus for so-called Volterra processes, i.e., processes which are defined as the stochastic integral of a time-dependent kernel with respect to a standard Brownian motion. For these processes which are natural generalization of fractional Brownian motion, we construct a stochastic integral and show some of its main properties: Regularity with respect to time and kernel, transformation under an absolutely continuous change of probability, possible approximation schemes and Itô formula.