Temporal models of innovation diffusion

New models for describing the time pattern of the spread of innovations through a population are presented. In general, most existing temporal models of innovation diffusion are binomial in the sense that they tacitly assume that the population can be divided, without loss of generality, into adopters and potential adopters. This paper discusses the limitations of the binomial models and presents the development and applications of a model where the potential adopter population is dynamic in nature. The possibility is considered of developing an analytical framework to study innovation diffusion processes where the population had polynomial dimensions-polynomial in the sense that it comprises members who are adopters, rejecters, disapprovers, and uncommitted. Some stochastic models to complement the most commonly used deterministic binomial models were developed. A few case studies are included.