Robust kalman filter for high-frequency financial data

Abstract

The robust recursive algorithm for the parameter estimation and the volatility prediction in GARCH models is proposed. The suggested technique employs principles of robustified Kalman filter. It seems to be useful for (high-frequency) financial time series contaminated by additive outliers. In particular, it can be effective in the risk control and regulation when the prediction of volatility is the main concern since it is capable of distinguishing and correcting outlaid bursts of volatility. This conclusion is confirmed by simulations and real data examples.