Case Article—Forecasting Beer Demand at Anadolu Efes

In this case study, the students are required to build a multiple linear regression model that explains the monthly beer demand in Turkey to help Efes Beverage Group in its beer demand predictions. The students are expected to study the factors affecting the beer demand, work on sets of raw data to transform them into relevant variables for the model, select the variables that are meaningful in explaining the demand, test the assumptions of regression, and validate their models, using variable transformations if necessary to obtain a valid model. They are also required to predict the future demand by the help of their models. They should try different scenarios of uncontrollable predictor variables and test the effects of changes in price-related controllable predictor variables. There are alternative approaches the students may take in carrying out various parts of the case.