Financial Options Pricing Using the MKPF Algorithm

A mixture Kalman Particle Filter (MKPF) based options pricing method is proposed. The MKPF algorithm uses the unscented Kalman filter (UKF) and the extended Kalman filter (EKF) as proposal distribution to generate the importance sampling density. Each particle is firstly updated by the UKF and obtains a state estimation. Thereafter, this estimation is used as the prior of the EKF, in which the particle is updated again to gain the final estimation of the state. We use the classical B-S model in the experiment aiming at evaluating the performance of the newly proposed method and other existing algorithms. The experimental results show that the MKPF outperforms other algorithms.