Comparison of the Predictability of a Neural Network with Retropropagation with Those using Linear Regression, Logistic and A.I.D. Methods for Direct Marketing Scoring
Date
1998
Dewey
Marketing direct
Sujet
empirical results; direct marketing; scoring; algorithm; bio mimetic; neural networks
In comparison with other usual statistical methods (MCO, logistic regression, discriminant analysis, AID), advantages of neural network with backpropagation are numerous and well known (non linear effects, distribution flee variables, low sensibility to outliers or missing variables). However, implementation and efficiency have not yet received a strong interest. The paper reviews comparative analyses and presents results obtained for prediction of a behaviour in Fund raising.
