Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/55186 
Year of Publication: 
2008
Series/Report no.: 
Working Paper No. 634
Publisher: 
Queen Mary University of London, Department of Economics, London
Abstract: 
Models based on economic theory have serious problems at forecasting exchange rates better than simple univariate driftless random walk models, especially at short horizons. Multivariate time series models suffer from the same problem. In this paper, we propose to forecast exchange rates with a large Bayesian VAR (BVAR), using a panel of 33 exchange rates vis-a-vis the US Dollar. Since exchange rates tend to co-move, the use of a large set of them can contain useful information for forecasting. In addition, we adopt a driftless random walk prior, so that cross-dynamics matter for forecasting only if there is strong evidence of them in the data. We produce forecasts for all the 33 exchange rates in the panel, and show that our model produces systematically better forecasts than a random walk for most of the countries, and at any forecast horizon, including at 1-step ahead.
Subjects: 
exchange rates
forecasting
Bayesian VAR
JEL: 
C53
C11
F31
Document Type: 
Working Paper

Files in This Item:
File
Size
838.38 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.