Minimum distance estimation of GARCH(1,1) models

A distribution free approach to the estimation of GARCH(1, 1) models is presented. More specifically, the proposed method relies on a Minimum Distance Estimator (MDE) based on the autocovariance function of the squared observations. The asymptotic properties of the estimator are studied giving conditions for its consistency and asymptotic normality while its finite sample efficiency is assessed by means of a simulation study. Finally the proposed estimation method is applied to a time series of hourly returns on the FIFSE100 index futures. (c) 2005 Elsevier B.V. All rights reserved.