Why are forecasts of inflation from VAR models so much worse then their forecasts of real variables? This paper documents that relatively poor performance, and finds that the price equation of a VAR model fitted to U.S. postwar data is poorly specified. Statistical work by other authors has...

An earlier version of this paper, "The Business Cycle and Economic Forecasting," was presented to the Western Economic Association in July 1986, and to the Federal Reserve System Research Committee on Business Analysis in November 1986.Macroeconomic forecasts are traditionally stated as point...

For much of the last four decades, leading analysts have used large Keynesian macroeconomic models to prepare macroeconomic forecasts. More recently, VAR models have become a popular alternative. Despite their usefulness in preparing unconditional forecasts, VAR models are unsuitable for policy...

A growing disenchantment with conventional economic models has resulted in increased interest in forecasting with vector autoregressive (VAR) models. In this article, Roy H. Webb develops a statistical procedure for determining the best configuration of explanatory variables in the equations of...

Thirty years ago it appeared that the best strategy for improving economic forecasts was to build bigger, more detailed models. As the cost of computing plummeted, considerable detail was added to models and more elaborate statistical techniques became feasible. Yet dissatisfaction with...