Forecasting from others’ experience: Bayesian estimation of the generalized Bass model

A Sustainable Road Transport Decarbonisation: The Scenario Analysis of New Energy Vehicle in China

“…With the rapid evolution of statistical theory and computer technologies, academics tend to use various intelligent algorithms to estimate coefficients of Bass models or generalised Bass models such as machine learning [ 42 ], sequential quadratic programming algorithms [ 43 ], and genetic algorithms [ 44 ], etc. The genetic algorithm, one of the frequently used algorithms, can be easily parallelised and has strong robustness and global search capabilities, so in this study, the genetic algorithm was chosen to estimate coefficients of the generalised Bass model.…”

A review of Bayesian dynamic forecasting models: Applications in marketing

“…Bayesian estimation of this class of models can be seen in Ramírez-Hassan and Montoya-Blandón. 28 The logistic model is slightly modified by Bass et al 27 including marketing components as the price and advertising effects, the coefficients of innovation and imitation, and the involved population. The dynamic version of Bass's generalized models (Bass et al 27 ) can be easily handled as a member of the wide class of dynamic growth curve models, that also includes the logistic, Gompertz, and modified exponential models (Migon and Gamerman 29 ).…”

“…The application of this class of models goes beyond marketing problems and can describe the evolution of an epidemic, retail services, industrial technologies, consumption of durable goods, and so forth. Bayesian estimation of this class of models can be seen in Ramírez‐Hassan and Montoya‐Blandón 28 . The logistic model is slightly modified by Bass et al 27 including marketing components as the price and advertising effects, the coefficients of innovation and imitation, and the involved population.…”

“…The previous model was extended by Bass et al 27 allowing it to include factors such as price growth rate and advertisement. As already mentioned, Ramírez‐Hassan and Montoya‐Blandón 28 recently proposed a Bayesian inference approach for the generalized Bass model. Although the Bayesian approach has several fundamental aspects to guide marketers, such as prior specifications by marketing experts and predictive distributions to support the decision making, taking into account the uncertainty, there are some drawbacks to this model that can be easily solved by the dynamic generalized exponential growth models (DGEGM) introduced by Gamerman and Migon 48 and Migon and Gamerman 29 .…”

Food Price Elasticities for Policy Interventions: Estimates from a Virtual Supermarket Experiment in a Multistage Demand Analysis with (Expert) Prior Information

“…The former reflects our prior beliefs about the magnitude of the coefficient and the latter reflects the strength of this belief with a large V corresponding to less weight on the prior belief. This can be considered a 'subjective empirical' Bayesian approach where informative priors are built based on previous studies to improve precision of posterior estimates due to shrinking sample and prior information (e.g., Ramírez-Hassan & Montoya-Bland ón, 2020).…”