We propose a new method for multivariate forecasting which combines the Generalized Dynamic Factor Model (GDFM) and the multivariate Generalized Autoregressive Conditionally Heteroskedastic (GARCH) model. We assume that the dynamic common factors are conditionally heteroskedastic. The GDFM,...

We use futures instead of forward rates to study the complete maturity spectrum of the forward premium puzzle from two days to six months. At short maturities the slope coefficient is positive, but these turn negative as the maturity increases to the monthly level. Futures data allow us to...