Parameter Estimations of Heston Model Based on Consistent Extended Kalman Filter

Heston model is widely applied to financial institutions, while there still exist difficulties in estimating the parameters and volatilities of this model. In this paper, the pseudo Maximum Likelihood Estimation and consistent extended Kalman filter (PMLE-CEKF) are implemented synchronously to estimate the Heston model. For parameter estimations, PMLE for the state equation and the measurement equation of the Heston model are conducted independently. For volatility estimations, the consistent extended Kalman filter (CEKF) algorithm is introduced to ensure the volatility to be well evaluated. Additionally, the estimation results of the Heston model are compared between PMLE-CEKF and PMLE-EKF algorithm. The numerical simulations illustrate that PMLE-CEKF algorithm works more efficiently than PMLE-EKF algorithm. Application of the PMLE-CEKF to S&P 500 shows the utility of the proposed algorithm. (C) 2017, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.