A Statistical Methodology for Specifying Neural Network Models: Application to the Identification of Cross-Selling Opportunities
Examining the recent applications of neurotechnology in the marketing field, one realises that the focus tends to be on comparing the predictive performance of neural networks to that of statistical models. The question addressed in most studies is “are neural networks better than statistical techniques ?”, and the published results appear to be both encouraging and discouraging (e.g. Furness 1995, Ripley 1994).
KeywordsNeural Network Logit Model Software Suite Hide Unit Constructive Algorithm
Unable to display preview. Download preview PDF.
- Aldrich J, Nelson F. Linear probability, logit, and probit models. Sage Publications, 1984.Google Scholar
- Bent, Y, Merunka D. Modelling brand choice with the multinomial logit model and neural networks: a comparison of methods and results. Proceedings of the Second International Meeting on Neural Approaches in Economics and Management Science ( ANSEG) Poitiers, France, 1995.Google Scholar
- Burgess AN. Methodologies for neural network systems. Proceedings of the Henry Stewart Conference on Neural Networks in Marketing, London, U.K. 1995.Google Scholar
- Fahlman S, Lebiere C. “The cascade-correlation learning architecture.” In Advances in Neural Information Processing Systems 2, D.S.Touretzky ed., Morgan Kaufmann, Los Altos CA: 1990; 524–532.Google Scholar
- Furness P. “Neural networks for data-driven marketing.” In Intelligent Systems for Finance and Business, S.Goonatilake and P.Treleaven eds., 1995; 73–96.Google Scholar
- Ripley BD. Neural networks and related methods for classification. Journal of the Royal Statistical Society, B, 1994; 56, 3: 409–456.Google Scholar
- Rumelhart DE, McClelland JL. Parallel Distributed Processing. MIT Press, 1986.Google Scholar
- Verley G, Asselin de Beauville JP. Multilayer Perceptron Learning Control. Proceedings of EURO-PAR `96, Ecole Normale Superieure, Lyon, August 1996; 27–29.Google Scholar