A0349
Title: Recursive estimation of multivariate GARCH models
Authors: Tomas Cipra - Charles University, Prague (Czech Republic) [presenting]
Abstract: Recursive algorithms for the parameter estimation and the volatility prediction in the multivariate GARCH models are suggested that seem to be useful for various financial time series, in particular for high frequency log returns which are (conditionally) correlated. These models are routinely estimated by computationally complex off-line estimation methods, e.g. by the conditional maximum likelihood procedure. However, in many empirical applications (especially in the context of UHF financial data) it seems necessary to apply on line methods which are numerically more effective to calibrate and monitor such models. Therefore, one (i) derives on-line estimation algorithms applying general recursive identification instruments for such models, and (ii) examines these methods by means of simulations and an empirical application.
