This book is the first of two volumes on macroeconomic forecasting. Its objective is to provide a formal analysis of the models, procedures and measures of economic forecasting with a view to improving forecasting practice. The first volume sets the scene, focusing on forecasting when the underlying process can be described by a stationary representation, perhaps after differencing the data. Our companion volume, entitled The Zeuthen Lectures on Economic Forecasting, will discuss forecasting in the presence of deterministic non-stationarities. Both volumes are the outcomes of Lecture Invitations, the first – the Marshall Lectures – at Cambridge University; the second at Copenhagen University. Of course, the choice of topics for these lecture series was conditioned by our on-going research interests. Our focus on forecasting began in 1991 when David Hendry was invited to act as a Special Adviser to the House of Commons Select Committee on the Enquiry into Official Economic Forecasting, and found a dearth of theory to explain the systematic mis-forecasting that had occurred in the mid and late 1980s. The literature on statistical forecasting also argued that there was a marked discrepancy between theory and practice: see, for example, Fildes and Makridakis (1995) and Makridakis (1982), among others. Thus, we were delighted to take up the implicit challenge of developing a theory of economic forecasting with practical implications.

The two volumes seek to address the key problems facing a theory of forecasting in macroeconomics based on empirical econometric models.