Foodservice Forecasting: Differences in Selection of Simple Mathematical Models Based on Short- Term and Long-Term Data Sets

Publisher Website
Google Scholar
Abstract
This study developed and evaluated mathematical (time-series) forecasting models to predict restaurant covers. The purpose of the study was to determine if model selection would differ for short-term and long-term data sets. In both the short- term and long-term studies, deseasonalized data modeled best. Therefore, daily seasonal differences account for a large portion of the demand variance, and the effect should be included in the forecasting model.