Comparative Analysis of Innovation Diffusion Models: Empirical Results and Predictive Performance on Russian Mobile Phone Propagation Data

Abstract

Abstract This article introduces a new model that describes the innovation diffusion and is an extension of the well-known logistic model to the case when a diffusion process has a more complex structure. Time series data of mobile phone subscribers for Russian Federation during 2000-2018 are examined to compare the performance of the proposed model with the well-known innovation diffusion models (the Gompertz, Logistic, Bass models) and the time-series autoregressive moving average (ARMA) model, one of the most popular forecasting models. Empirical results show that the extended logistic model outperforms the other models and the proposed model has the best characteristics on real data for the Russian mobile communications market.