Small-scale VARs are widely used in macroeconomics for forecasting US output, prices, and interest rates. However, recent work suggests these models may exhibit instabilities. As such, a variety of estimation or forecasting methods might be used to improve their forecast accuracy. These include...

The relationships between real wages, output per capita, inflation and unemployment in Italy between 1970 and 1994, are modelled using a cointegrated vector autoregression. There is evidence of a change in the underlying equilibria and in the dynamic evolution of the variables, probably...

The relationship between wages, prices, productivity, inflation and unemployment in Italy, Poland, and the UK between the 1960s and the early 1990s is modelled as a cointegrated vector autoregression subject to regime shifts. For each of these economies there is clear evidence of a change in the...

Models based on economic theory have serious problems 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...

We show how the use of panel data methods such as those proposed in single equations by Kao and Pedroni or in systems by Larsson and Lyhagen to investigate economic hypotheses such as purchading power pariety or the term structure of interest rates may be affected by the existence of cross-unit...

This paper analyses two features of concern to policy-makers in the countries of the prospective European Monetary Union: The solvency of their governments finances; and the accuracy of fiscal forecasts.