DETERMINING THE SALIENCY OF INPUT VARIABLES IN NEURAL-NETWORK CLASSIFIERS

This paper examines a measure of the saliency of the input variables t hat is based upon the connection weights of the neural network. Using Monte Carlo simulation techniques, a comparison of this method with th e traditional stepwise variable selection rule in Fisher's linear clas sification analysis (FLDA) is made. It is found that the method works quite well in identifying significant variables under a variety of exp erimental conditions, including neural network architectures and data configurations. In addition, data from acquired and liquidated firms i s used to illustrate and validate the technique. (C) 1997 Elsevier Sci ence Ltd.