CMStatistics 2020: Start Registration
View Submission - CMStatistics
Title: An asymptotic approximation for the extended Bass diffusion model Authors:  Ringo Thomas Tchouya - IMSP (Benin)
Sophie Dabo - University of Lille (France) [presenting]
Stefano Nasini - IESEG School of Management (France)
Abstract: The continuous-time dynamic problem of diffusion in bipartite networks is studied, and a model is proposed in which an influencer node is responsible for the propagation of a binary state towards multiple independent populations. We deduce an asymptotically correct approximation scheme for the underlying differential equation, that allows embedding the model into computationally treatable estimation approach. This supports the estimation of the unknown parameters of this model, using observed time series data. A comprehensive numerical study is presented to assess the validity of the proposed approach.