Forecasting performance of Bayesian VEC-MSF models for financial data in the presence of long-run relationships

Abstract The paper is focused on comparing the forecasting performance of two relatively new types Vector Error Correction - Multiplicative Stochastic Factor (VEC-MSF) specifications: VEC-MSF with constant conditional correlations, and VEC-MSF-SBEKK time-varying correlations. For sake comparison, random walks, vector autoregressions (VAR) covariance matrix, VAR-SBEKK models are also considered. Based daily quotations three exchange rates: PLN/EUR, PLN/USD, EUR/USD, where cointegrating may be assumed to known a priori , we show that in econometric it can more important allow for cointegration relationships than matrix.