We apply hazard function models to marketing data on household purchases of disposable diapers to calibrate repeat-purchase and brand-switching behaviors. In the model we allow for household-specific unobserved traits. Hazard function models with fixed-effects heterogeneity specification are computationally feasible. We directly compare the results of fixed- and random-effects heterogeneity specifications in the context of parametric and nonparametric duration-dependence functions. We find that a rigorous test fails to reject the random-effects specification in favor of the fixed-effects specification. The unimodal gamma distribution in the random-effects specification adequately captures unobserved heterogeneity.
No takes yet. Share an insight, caveat, or question.
Gönül et al. (1993) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: