Forecasting and analysis of marketing data using neural networks: a case of advertising and promotion impact

The paper explores the use of neural networks in analyzing the effects of advertising and promotion in the retail business. In retailing, one of the key problems is the optimal allocation of advertising expenses and the forecasting of the total sales levels across a wide product range where significant cross-effects among products are likely. Various statistical and econometric methods can be used for such analyses. However, neural networks can offer an alternative and potentially beneficial approach to handling this ill-structured problem. The study evaluates their performance in studying the impact of advertising and promotion on sales. The results reveal that the predictive quality of neural networks depends on the different frequency of data observed i.e. daily or weekly-data models, and the specific learning algorithms used. The study shows that neural networks are capable of capturing the nonlinear aspects of complex relationships in non-stationary data. By performing sensitivity analysis, neural networks can potentially single out important input variables, thereby making it useful for scenario development and practical use