Using only aggregate sales data, the model we propose decomposes the diffusion processes of the respective technological generations and tests if different technological generations have different diffusion parameters. It also estimates the location of the generational transition from the old to the new technology. We develop a routine to test whether the maturation point of the old generation occurs before or after the transition to a new technological generation. Finally, we show that, when the aggregate sales data are generated by multiple technological generations, our model does better in forecasting than a single-regime Bass model.

diffusion modeling, regime-switching models, technological generations
Statistical Decision Theory; Operations Research (jel C44), Business Administration and Business Economics; Marketing; Accounting (jel M), Marketing (jel M31), Technological Change: Choices and Consequences; Diffusion Processes (jel O33)
Erasmus Research Institute of Management
hdl.handle.net/1765/199
ERIM Report Series Research in Management
Copyright 2002, P.H. franses, S. Stremersch, This report in the ERIM Report Series Research in Management is intended as a means to communicate the results of recent research to academic colleagues and other interested parties. All reports are considered as preliminary and subject to possibly major revisions. This applies equally to opinions expressed, theories developed, and data used. Therefore, comments and suggestions are welcome and should be directed to the authors.
Erasmus Research Institute of Management

Franses, Ph.H.B.F, & Stremersch, S. (2002). Modeling Generational Transitions from Aggregate Data (No. ERS-2002-49-MKT). ERIM Report Series Research in Management. Erasmus Research Institute of Management. Retrieved from http://hdl.handle.net/1765/199