Beyond Accuracy: Comparison of Criteria Used to Select Forecasting Methods

Abstract

Two studies were conducted to examine expert opinions of criteria used to select forecasting techniques. In Study One, while accuracy was a dominant criterion, the ratings of five of thirteen criteria varied by the role of the forecaster. Researchers rated accuracy relatively higher than did practitioners, educators or decision-makers. Decision makers rated implementation-related criteria, such as ease criteria, relatively higher than the other groups. In Study Two, forecasting experts significantly varied their ratings on six of seven criteria according to situations. Other criteria were often as important or more important than accuracy, especially when the situation involved making many forecasts. In general, there was much agreement across roles and across situations that accuracy was the most important criterion, but other criteria were rated as being almost as important. In particular, factors related to implementation, such as ease of interpretation and ease of use, were highly rated.

Keywords: Expert opinion, Forecaster's role, Forecast situation, Implementation