Abstract
This research has two objectives. The first is to develop a conceptual neural network for studying manufacturer-distributor cooperation in the new product development (NPD) process and to compare the neural network directly with the traditional multiple regression. The second objective is to examine the relative importance of the antecedents of manufacturer-distributor cooperation. Data from 295 U.S. manufacturing firms are used to test the neural models. The study demonstrates that neural network analysis is a good method predicting manufacturer-distributor cooperation in the NPD process. The results also show that the ranking of antecedents of manufacturer-distributor cooperation from most to least important is: relative dependence, shared values, communication, commitment, and trust. Implications for NPD managers are offered at the end of the paper.