A Logistic Neural Network Approach to Extended Warranty

Abstract

An extended warranty, sometimes called a service agreement, a service contract, or a maintenance agreement, is a prolonged warranty offered to consumers. Studying the extended warranty is extremely important for business investors and policymakers for effective warranty planning. However, measuring, forecasting and tracking the global diffusion of extended warranty have not been researched. This study uses model based on the knowledge of traditional diffusion theory as well as artificial neural networks. Additionally, it integrates the two into a hybrid model in order to study extended warranty growth. A count of green warranty can be used as a reliable measure of extended warranty growth in all the models. Our study demonstrates that a logistic Neural Network model, if properly calibrated, can create a very flexible response function to forecast the extended warranty amounts.