ABSTRACT

This paper shows how the recently developed fractional FFT algorithm (FRFT) can be used to retrieve option prices from the corresponding characteristic functions. The FRFT algorithm has the advantage of using the characteristic function information in a more efficient way than the straight FFT. Typically, therefore, fewer function evaluations are needed and substantial savings in computational time can be made. Two experiments, based on the stochastic volatility and the variance–gamma models, illustrate the benefits of using the fractional version of the FFT and show that option prices can be delivered up to 45 times faster without substantial loss of accuracy in the results.