Forecasting UK In ation: Empirical Evidence on Robust Forecasting Devices

Forecasting in ation is fundamental to UK monetary policy, both for policy-makers and private agents. However, forecast failure is prevalent with naive devices often outperforming the dominant congruent in-sample model in forecasting competitions. This paper assesses evidence for UK annual and quarterly in ation using the theoretical framework developed by Clements and Hendry (1998, 1999) to explain the empirical ndings. We build both single equation and multivariate equilibrium correction models of in ation using the automatic model selection algorithm, PcGets, and use these models along with various transformations of the models to forecast UK in ation over the period 1997-2003. Robust forecasting devices do prove useful in forecasting macroeconomic time series and they often outperform econometric models, both when there are structural breaks in the data and when the underlying process appears to be stable but with breaks in the explanatory variables. Increasing the information set does lead to improvements in forecasting performance suggesting that disaggregation can yield bene ts. Finally, it is observed that much of the forecast error in the structural models is driven by the deterministic terms. Breaks in the mean of the cointegrating vector or the growth rate of the system will cause forecast `failure' and results show how sensitive forecasts are to errors in these terms.