This book describes the creation of a simulation model that is designed to serve as an artificial test market. The market is sufficiently realistic to permit it to be used for experimenting with a wide range of alternative marketing programs for a manufacturer of a branded, frequently purchased food product.
The simulator includes characterizations by brand and size of: (1) the rates of retailer promotions (such as retail price discounts, newspaper advertising, displays, signs and banners, special offers, and premiums) as a function of the promotional allowances offered by competing manufacturers; (2) the response of costumers in terms of sales volume to the aforementioned retailer's promotional activities; and (3) the transformation of these volume movement figures into estimates of manufacturer profitability. The simulation model is designed to take advantage of modern time-sharing computation facilities and permit virtually instantaneous assessment of alternatives, and to encourage the combining of executive judgments on current market conditions and probable retailer responses with historical data.
In order to develop the component of the stimulator which relates retailer promotional activities to consumer volume movement, econometric methods are used to analyze a remarkably complete set of data in such a way that the reader may follow the author's thought processes as they work through complex quantitative techniques. The econometric model that they develop is a multiple-equation model. The use of this type of model is relatively new in marketing, although econometricians have been familiar with these techniques for a number of years. The present study develops and discusses the special statistical methods required to estimate the parameters of such a model and postulates a specific model linking the demand for a given brand and size to the promotional activities of all the brand-size combinations present in the marketplace for the product in question.
