Forecasting Sales: A Model and Some Evidence from the Retail Industry

Abstract

This paper presents a sales forecasting model and tests the model on a sample of firms in the retail industry. The model distinguishes between sales growth due to an increase in the number of sales-generating units (e.g. opening new stores) and growth due to an increase in the sales rate at the existing units (e.g. the comparable store growth rate). The model accommodates different trends in the sales rates, allowing new stores to earn more or less than existing stores, perhaps because new stores are different sizes than existing stores or may take either a long time to reach maturity or alternatively enjoy an early “fad” status. We show how to use the historical series of sales, stores and comparable store growth rates to estimate the sales rates on new stores and on existing stores. The model uses only a few years of firm-specific, publicly available information, yet generates in-sample forecast errors of less than two percent of sales, generates out-of-sample forecast errors that are comparable to analyst revenue forecasts, and when used with analyst forecasts, adds significant incremental information.