Pricing of Swing Options in a Mean Reverting Model with Jumps

We investigate the pricing of swing options in a model where the logarithm of the spot price is the sum of a deterministic seasonal trend and an Ornstein-Uhlenbeck process driven by a jump diffusion. First we calibrate the model to Nord Pool electricity market data. Second, the existence of an optimal exercise strategy is proved, and we present a numerical algorithm for computation of the swing option prices. It involves dynamic programming and the solution of multiple parabolic partial integro-differential equations by finite differences. Numerical results show that adding jumps to a diffusion may result in 2-35% higher swing option prices, depending on the moneyness and timing flexibility of the option.