Choosing among diffusion models: some empirical evidence
Author: Parker, Philip M. INSEAD Area: MarketingIn: Marketing Letters, vol. 4, no. 1, 1993 Language: EnglishDescription: p. 81-94.Type of document: INSEAD ArticleNote: Please ask us for this itemAbstract: A number of econometric diffusion models have been suggested in the marketing literature to help researchers and managers better understand the implications of both short-run and long-run adoption processes. Is there a "best" diffusion model ? This paper responds to calls for an integration and evaluation of both parcimonious and more complex flexible-form diffusion models. The study focuses on models which find their origins in the mixed-influence model proposed by Bass (1969). First purchase data collected for 19 durable goods are used to evaluate 24 alternative diffusion models. Two approaches are used for the systematic comparison. The first is a meta-analytic study of all categories pooled together. This procedure allows one to deduce, for example, which among all three-parameter diffusion models are likely to provide, on average, the highest explanatory power. The second is an econometric apporach which determines, for each category, the best fitting model based on likelihood ratio testsItem type | Current location | Call number | Status | Date due | Barcode | Item holds |
---|---|---|---|---|---|---|
![]() |
Europe Campus | Available | BC003939 |
Ask Qualtrics
A number of econometric diffusion models have been suggested in the marketing literature to help researchers and managers better understand the implications of both short-run and long-run adoption processes. Is there a "best" diffusion model ? This paper responds to calls for an integration and evaluation of both parcimonious and more complex flexible-form diffusion models. The study focuses on models which find their origins in the mixed-influence model proposed by Bass (1969). First purchase data collected for 19 durable goods are used to evaluate 24 alternative diffusion models. Two approaches are used for the systematic comparison. The first is a meta-analytic study of all categories pooled together. This procedure allows one to deduce, for example, which among all three-parameter diffusion models are likely to provide, on average, the highest explanatory power. The second is an econometric apporach which determines, for each category, the best fitting model based on likelihood ratio tests
Digitized
There are no comments for this item.