Abstract
There is a fundamental difference between stochastic consumer behaviour models and most models described up to now in that,
‘a stochastic model is a model in which the probability components are built in at the outset rather than being added ex post facto to accomodate discrepancies between predicted and actual results’ (Massy, Montgomery and Morrison, 1970, p. 4).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Nicosia (1972) calls these models behaviouristic. The label ‘behavioural’ is reserved for models where other factors are explicitly considered.
t ∈ T means that t is an element (or member) of the set T, or, in other words, is a short hand notion for writing t belongs to T.
In other words, what is the total demand in a specified interval of time.
In this book, we do not consider the question which stochastic process describes the purchase data best. The interested reader is referred to Morrison (1966), Montgomery (1969), and Wierenga (1974).
Although it is not our purpose to give a detailed study of brand choice models some introduction is inevitable. For an extensive review, see, for example, Massy, Montgomery, Morrison (1970) and Wierenga (1974). Introductory treatments are Montgomery and Urban (1969, pp. 53–93), Simon and Freimer (1970, Ch. 10). For a treatment of Markov models we refer to Leeflang (1974, Ch. 3, 7).
The latter name is used by Bass (1974).
For an example of estimation of transition probabilities from panel data, see Telser (1962a).
See also Telser (1962b, 1963).
See also Section 4.1.
See also the application from Lambin (1970), discussed above.
See also Jones (1973). In the same way a homogeneous Bernouilli model can be transformed into a heterogeneous Bernouilli model, as is shown by Massy, Montgomery and Morrison (1970, pp. 59–78).
See in this resepct Ehrenberg’s (1965) pessimistic view on the possibilities of Markov chain applications in brand choice processes.
No authors listed.
A fact confirmed in an empirical context by Colard (1975), whose basic assumption was a heterogeneous multinomial probability model.
See Starr (1975, pp. 199–200). Herniter (1974) compared the Hendry Model with his own entropy model (Herniter, 1973). For an axiomatic definition and an extensive treatment of entropy in an economic context, see Theil (1967).
Brand 3 is only sold in cups.
We remind the reader that the illustration is hypothetical and that the conclusions may, therefore, perhaps be somewhat overstated.
We can leave out the brand index since in learning models one generally considers only two brands.
See Kuehn (1962), and Herniter and Howard (1964).
Again it is assumed that g = l.
See, for example, Carman (1966), McConnell (1968).
See, for example, Kotier (1971, p. 474).
The variable v = the average amount purchased by all individuals is replaced in this model by λ k the average amount purchased by an individual k, k = 1,…, N.
troll is an interactive estimation and simulation package developed by the Computer Research Center, National Bureau of Economic Research, Inc. For a description of the estimation method, see Eisner and Pindyck (1972).
On poisson-type purchase models, see equations (10.31) to (10.33).
For a more thorough discussion about the way a normative model could be built using a consumer behaviour model and a response model, we refer to Leeflang (1974).
Lilien (1976) is another example of allocating a retail outlet building budget across market areas.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1978 H. E. Stenfert Kroese B. V.
About this chapter
Cite this chapter
Naert, P.A., Leeflang, P.S.H. (1978). Stochastic consumer behaviour models. In: Building Implementable Marketing Models. Springer, Boston, MA. https://doi.org/10.1007/978-1-4615-6586-4_10
Download citation
DOI: https://doi.org/10.1007/978-1-4615-6586-4_10
Publisher Name: Springer, Boston, MA
Print ISBN: 978-90-207-0674-1
Online ISBN: 978-1-4615-6586-4
eBook Packages: Springer Book Archive