Method of moments estimation of GO-GARCH models

We propose a new estimation method for the factor loading matrix in generalized orthogonal GARCH (GO-GARCH) models. The method is based on the eigenvectors of a suitably defined sample autocorrelation matrix of squares and cross-products of the process. The method can therefore be easily applied to high-dimensional systems, where likelihood-based estimation will run into computational problems. We provide conditions for consistency of the estimator, and study its efficiency relative to maximum likelihood estimation using Monte Carlo simulations. The method is applied to European sector returns, and to the correlation between oil and kerosene returns and airline stock returns.