1 Estimation of Vector Error Correction Model with GARCH Errors Koichi Maekawa Hiroshima University of Economics Kusdhianto Setiawan Hiroshima University of Economics and Gadjah Mada University Abstract The standard vector error correction (VEC) model assumes the iid normal distribution of disturbance term in the model. This paper extends this assumption to include GARCH process. We call this model as VEC-GARCH model. However as the number of parameters in a VEC-GARCH model are large, the maximum likelihood (ML) method is computationally demanding. To overcome these computational difficulties, this paper searches for alternative estimation methods and compares them by Monte Carlo simulation. As a result a feasible generalized least square (FGLS) estimator shows comparable performance to ML estimator. Furthermore an empirical study is presented to see the applicability of the FGLS. 1. Introduction Vector Error correction (VEC) model is often used in econometric analysis and estimated by maximum likelihood (ML) method under the normality assumption. ML estimator is known as the most efficient estimator under the iid normality assumption. However there are disadvantages such that the normality assumption is often violated in real date, especially in financial time series, and that ML estimation is computationally demanding for a large model. Furthermore in our experience of empirical study error terns in VEC model often show a GARCH phenomenon, which violates iid assumption. To overcome these disadvantages and to reduce computational burden of ML estimator it may be worthwhile to reconsider the feasible generalized least square (FGLS) estimator instead of ML estimator (MLE) because FGLS method is relatively free from the distributional assumptions and ease computational burden. The purpose of this paper is to examine the finite sample properties of FGLS estimator in VEC-GARCH model by Monte Carlo simulation and by real data analysis of the international financial time series. The paper is organized as follows: Section 2 briefly surveys the multivariate GARCH (MGARCH hereafter) model; Section 3 describes VEC representation of the vector autoregressive (VAR) model; Section 4 presents a VEC- GARCH model and shows that this model can be estimated by FGLS within the framework of the seemingly unrelated regression (SUR) model; Section 5 examines the performance of FGLS by Monte Carlo simulation; Section 6 presents an empirical application of VEC-GARCH model and shows the applicability of FGLS; finally Section 7 gives some concluding remarks. 2. Multivariate GARCH Multivariate GARCH model has been developed and applied in financial econometrics and numerous literature were published. The recent development in this