Modeling diffusion of innovation with cellular automata

Abstract

Due to the lack of individual-level data, CA models that are based on individual-level theory are difficult to be validated quantitatively. In this paper, we present an approach by which we can derive the parameters of CA model from aggregate-level data so that the presented model can be used for the prediction of Internet users' quantity in the future. The result of simulation suggests that the forecasting precision of the presented model is nearly equal to the aggregate-level model (Bass model). Moreover, this model can reproduce the pattern of Internet diffusion in the aggregate-level by the interacting of individual level units, which could not be educed by the Bass model.