Economic Forecasting: Some Lessons from Recent Research

We describe a general theoretical framework against which recent results in economic forecasting can be judged, including explanations for the findings of forecasting competitions, the prevalence of forecast failure, and the role of causal variables. We compare this framework to a previous formulation which was silent on the very issues of most concern to the forecaster, then describe ten aspects which our approach illuminates, and draw out their implications for model selection. Finally, we discuss ten areas where research is needed to clarify empirical findings that still lack theoretical explanations.