ANALYZING DURATION TIMES IN MARKETING - EVIDENCE FOR THE EFFECTIVENESS OF HAZARD RATE MODELS

Some statistical methods developed recently in the biometrics and econ ometrics literature show great promise for improving the analysis of d uration times in marketing. They incorporate the right censoring that is prevalent in duration times data, and can be used to make a wide va riety of useful predictions. Both of these features make these methods preferable to the regression, logit, and discriminant analyses that m arketers have typically used in analyzing durations. This paper is int ended to fulfill three objectives. First, we demonstrate how decision situations that involve durations differ from other marketing phenomen a. Second, we show how standard modeling approaches to handle duration times can break down because of the peculiarities inherent in duratio ns. It has been suggested in recent marketing articles that an alterna tive to these conventional procedures, i.e., hazard rate models and pr oportional hazard regression, can more effectively handle duration typ e data. Third, to investigate whether these proposed benefits are in f act delivered for marketing durations data, we estimate and validate b oth conventional and hazard rate models for household interpurchase ti mes of saltine crackers. Our findings indicate the superiority of prop ortional hazard regression methods vis-a-vis common procedures in term s of stability and face validity of the estimates and in predictive ac curacy.