| |
Abstract:
This paper contains a survey of the recent literature on the evaluation of direct marketing campaigns. We give an outline of the various stages included in such a campaign. Next, we review the statistical methods most frequently used and we review the general findings from using these methods.
direct marketing, evaluation, quantitative models, target selection
|
|
2.
|
|
|
Jedid-Jah Jonker Social and Cultural Planning Office Nanda Piersma Amsterdam School of Business - Academy for Economic Studies Rob Potharst Erasmus University Rotterdam - Department of Computer Science
|
| Posted: |
|
26 Aug 06
|
|
Last Revised:
|
|
26 Aug 06
|
|
106 (75,640)
|
|
|
| |
Abstract:
Direct marketing firms want to transfer their message as efficiently as possible in order to obtain a profitable long-term relationship with individual customers. Much attention has been paid to address selection of existing customers and on identifying new profitable prospects. Less attention has been paid to the optimal frequency of the contacts with customers. We provide a decision support system that helps the direct mailer to determine mailing frequency for active customers. The system observes the mailing pattern of these customers in terms of the well known R(ecency), F(requency) and M(onetary) variables. The underlying model is based on an optimization model for the frequency of direct mailings. The system provides the direct mailer with tools to define preferred response behavior and advises the direct mailer on the mailing strategy that will steer the customers towards this preferred response behavior.
Markov decision process, direct marketing, decision support system
|
| |
Abstract:
We analyse the demand for health care in The Netherlands. The goal of the analysis is to provide insight into the (determinants of the) size of the demand for the formal care provisions at a macro-level. In this paper we focus on the relationship between the use of the several care provisions. In particular, we examine how health and personal characteristics affect decisions about whether to seek care, and from whom. We consider eight available alternatives of care, namely (1) no care, (2) (private) informal home care, (3) (private) formal home care, (4) (public) household care, (5) (public) personal care, (6) (public) nursing care, (7) care in a rest home or (8) care in a nursing home. We assume that the set of available alternatives are combinations of three underlying choice dimensions: the decision to use care, the decision to stay at home and the choice of provision. If the person decides to stay at home, he can choose between private home care and public home care. Private care can be either informal or formal. Public care can be household care, personal care or nursing care. If someone chooses for an institutional living, (s)he can choose between care in a rest home and care in a nursing home. In such a choice setting, a nested multinomial logit model is appropriate. The data are derived from a match of two 2003-surveys. The first survey was held among independently living persons. Data were collected on personal characteristics, household composition, health status, and use of care provisions. The second survey contains information on residents of rest homes and residents of nursing homes. In total, we have observations on about 10.000 individuals. These surveys together provide a representative sample of the Dutch population in 2003.The picture that shows up is that a rest home is predominantly for elderly who get problems with taking care of themselves just because of old age, rather than because of illness. Females and couples have a lower probability of ending up in a rest or nursing home compared to males and singles. Income has a negative effect on the probability of receiving any care and also the price effects are significant. By allowing the price coefficients to vary over the choices, we try to capture some 'necessity' effects. The demand for one alternative is more price sensitive than the other because the necessity of care is greater for one alternative than for another. We find that the demand for care in a rest home is more price elastic than the demand for care in a nursing home. We have combined the estimated micro relations with predicted macro-numbers of the determinants to obtain macro predictions of the demand for several years. These predictions indicate that the demand for rest homes and nursing homes will increase faster than the growth of the population, whereas the demand for home care will be more or less equal to the growth of the population.
demand for care, nested logit model, forecasting
|