Stochastic models of consumer behaviour

Wagner, Udo; Taudes, Alfred

doi:10.1016/0377-2217(87)90189-5

Cited by 39 publications

(11 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The negative binomial distribution used below results from the homogeneous Poisson distribution by assuming that the conditional mean λ follows a Gamma distribution. A detailed substantiation of the negative binomial distribution for modeling heterogeneity in consumer behavior was previously provided by Ehrenberg (1959) as well as Wagner and Taudes (1987), for instance. In addition to this behavior-oriented motivation, there is also a more technical one.…”

Section: The Negative Binomial Regression Approachmentioning

confidence: 99%

Estimating aggregate consumer preferences from online product reviews

Decker

Trusov

2010

International Journal of Research in Marketing

264

View full text Add to dashboard Cite

Section: The Negative Binomial Regression Approachmentioning

confidence: 99%

Estimating aggregate consumer preferences from online product reviews

Decker

Trusov

2010

International Journal of Research in Marketing

264

View full text Add to dashboard Cite

“…choice, as has been pointed out in several reviews (see, e.g., Robertson and Hassarjiam, 1991;Wagner and Wright, 1994;Maurai, 1995).…”

mentioning

confidence: 87%

A multi‐attribute motivational model of consumer choice based on a new outranking relation

Tavares¹

2003

Int Trans Operational Res

View full text Add to dashboard Cite

Multiple models have been proposed to describe different patterns of consumer behavior, that is becoming more complex and diversified (so called ''postmodern consumer''), but little attention has been given to explore the potential of the approach based on pairwise comparisons. In this paper, a new model of consumer choice is developed considering that: the space of the perceptual attributes includes two subspaces; the motivational attributes which are responsible for the generation of a motivational drive or force; other relevant decisional attributes. The process of choice is modeled by an outranking relation that is appropriate to describe pairwise comparisons. An example is presented illustrating the applicability of this model to the choice of a sport car, based on the two groups of attributes. There is general agreement about the impossibility of developing an universal model for consumer choice but it is expected that this model can enrich the list of available candidates. Major advantages are: the explicit distinction between motivational and non-motivational attributes. These two groups play a different role in the process of consumer decision making; the description of the consumer choice in terms of successive pairwise comparisons; and the use of a new outranking relation appropriate to model the choice made by the consumer in terms of his motivation and avoiding the integration of multiple attributes through the usual weighting functions which are hardly confirmed by empirical evidences.

show abstract

“…The purchases of a consumer in successive periods follow a Poisson distribution with a certain long-run average . The purchases of a consumer follow a Poisson distribution in subsequent periods if a purchase tends to be independent of previous purchases (as is often observed) and a purchase occurs in such an irregular manner that it can be regarded as if random (see Wagner and Taudes [Wagner and Taudes, 1987] A -distribution of the in the population may have the following reason (see Ehrenberg [Ehrenberg, 1988, p. 259]): If for different products È É Ê Ë the average purchase rate of È is independent of the purchase rates of the other products, and È È ·É·Ê·Ë· µ is independent of a consumer's total purchase rate of buying all the products, then it can be shown that the distribution of must be . These independence conditions are likely to hold approximately in practice (see e.g.…”

Section: Ehrenberg´s Repeat-buying Theory and Bundles Of Information mentioning

confidence: 99%

Behavior-Based Recommender Systems as Value-Added Services for Scientific Libraries

Geyer-Schulz¹,

Hahsler²,

Neumann³

et al. 2003

Statistical Data Mining and Knowledge Discovery

View full text Add to dashboard Cite

Abstract. Amazon.com paved the way for several large-scale, behavior-based recommendation services as an important value-added expert advice service for online book shops. In this contribution we discuss the effects (and possible reductions of transaction costs) for such services and investigate how such a value-added service can be implemented in the context of scientific libraries. For this purpose we present a new, recently developed recommender system based on a stochastic purchase incidence model, present the underlying stochastic model from repeat-buying theory and analyze whether the underlying assumptions on consumer behavior hold for users of scientific libraries, too. We have analyzed the logfiles with approximately 85 million http-transactions of the web-based online public access catalog (OPAC) of the library of the Universität Karlsruhe (TH) since January 2001 and performed some diagnostic checks. A test prototype is already operational and is currently being evaluated. The recommender service will be fully operational within the library system of the Universität Karlsruhe (TH) by the end of June 2002.

show abstract

Stochastic models of consumer behaviour

Cited by 39 publications

References 34 publications

Estimating aggregate consumer preferences from online product reviews

Estimating aggregate consumer preferences from online product reviews

A multi‐attribute motivational model of consumer choice based on a new outranking relation

Behavior-Based Recommender Systems as Value-Added Services for Scientific Libraries

Contact Info

Product

Resources

About