ABSTRACT
Our aim in this chapter is not to try to fit a volatility smile exactly, but to add a measure of convexity to an existing skew. We develop a shifted-stochastic volatility model with zero correlation between the forwards and their stochastic volatility, in which the shift (rather than the absent correlation) is used to fit the skew. Independence of the yieldcurve and stochastic volatility drivers simplifies
the mathematics, and yields a model that when compared to Shifted BGM:
1. Has only a few more parameters and is only marginally more complex to calibrate using techniques already developed.
