Application of Kalman Filter to Stochastic Volatility Models of the Orstein‑Uhlenbeck Type

Barndorff‑Nielsen and Shephard (2001) proposed a class of stochastic volatility models in which the volatility process is the Ornstein‑Uhlenbeck process driven by a Levy process without gaussian component. Parameter estimation of these models is difficult because the appropriate likelihood functions do not have a closed‑form expression. The article deals with application of the Kalman filter technique for parameter estimation of such models. The method is applied to EUR/PLN daily exchange rate data. Empirical application is accompanied with simulation study to examine statistical properties of the estimators.