The Forecasting Canon: Nine Generalizations to Improve Forecast Accuracy

Abstract

Using findings from empirically-based comparisons, Scott Armstrong develops nine generalizations that can improve forecast accuracy. He finds that these are often ignored by organizations, so that attention to them offers substantial opportunities for gain. In this paper, Scott offers recommendations on how to structure a forecasting problem, how to tap managers' knowledge, and how to select appropriate forecasting methods.

Keywords: forecast method selection, domain knowledge, decomposition, judgmental bootstrapping, causal models, simulated interaction, combining forecasts

JEL Classification: CO, C2