Forecasting Restaurant Sales Using Multiple Regression And Box-Jenkins Analysis

Abstract

Several regression and Box-Jenkins models were used to forecast weekly sales at a small campus restaurant for Years 1 and 2.  Forecasted sales were compared with actual sales to select the three most promising forecasting models.  These three models were then used to forecast sales for the first 44 weeks of Year 3, and compared against actual sales.  The results indicate that a multiple regression model with two predictors, a dummy variable and sales lagged one week, was the best forecasting model considered.