Neural Networks for the Analysis and Forecasting of Advertising and.pdfVIP

Neural Networks for the Analysis and Forecasting of Advertising and Promotion Impact HeanLee Poh Jingtao Yao Teo Ja si c Scho ol of Computing National University of Singap ore Singap ore Republic of Singap ore Tel Fax email yao jtcompnusedusg First Draft June st Revision February nd Revision August Intel ligent Sysytems in Accounting Finance and management Vol No Neural Networks for the Analysis and Forecasting of Advertising and Promotion Impact Abstract A l locating advertising expenses and forecasting total sales levels are the key issues in retailing especial ly when many products are covered and signicant crosseects among products are likely Various statistical and econometric methods could be applied for such analyses We explore how wel l neural networks can be used in analyzing the eects of advertising and pr omotion on sales in this article The results reveal that the pr edictive quality of neural networks depends on the dierent fr equency of data observed ie daily or weekly data models and the specic learning algorithms used The study also shows that neural networks are capable of capturing the nonlinear aspects of complex relationships in nonstationary data By performing sensitivity analysis neural networks can p otent