Modelling the consumer's decision to replace durable goods: a hazard function approach

Abstract

This article analysis the consumer's durable good replacement decision using hazard models. In contrast to the typical limited dependent variable model often used in durable good demand studies, hazard models allow for much richer replacement. To illustrate the technique, a recursive system consisting of a regression equation and a hazard model is used to examine home heating system replacement decisions by residencial customers of a major southeasten US electric utility. The results indicate that overall system replacement rates decline over time, and that the probability of replacement for specific households depends negatively on the age of the head of household abd the availability of natural gas, and positively on system age and higher than expected household energy use.