The Mixed-Media Dirichlet Multinomial Distribution: A Model for Evaluating Television-Magazine Advertising Schedules

Abstract

Increasing awareness of the qualitative differences between television and magazine advertisements and the resulting inference that a response to a television ad is much different from a response to a magazine ad suggest the appropriateness of a model which jointly evaluates an advertising schedule including both television and magazines. The mixed-media Dirichlet multinomial distribution model proposed estimates the joint frequency distribution of exposure of such a mixed-media advertising schedule and evaluates response under a variety of alternative theoretical assumptions. An empirical test is performed to evaluate the accuracy of the model, and the model's performance is compared with that of models with different theoretical assumptions. The results have important implications for media planning, with a strong case being made for coordinated planning of the television and magazine advertising schedules.