Univ., Diss.--Bern, 2005.

Zugl. Diss. Bern, 1998 u.d.T.: Das Modell der Segmentintensität. / Zugl.: Diss. Bern, 1998 unter dem Titel "Das Modell der Segmentintensität" Literaturverz.

There are numerous ways of segmenting a market based on consumer survey data. We introduce bagged clustering as a new exploratory approach in the field of market segmentation research which offers a few major advantages over both hierarchical and partitioning algorithms, especially when dealing with large binary data sets: In the hierarchical step of the procedure the researcher is enabled to inspect if cluster structure exists in the data and gain insight about the number of clusters to extract. The bagged clustering approach is not limited in terms of sample size, nor dimensionality of the data. More stable clustering results are found than with standard partitioning methods (the comparative evaluation is demonstrated for the K-means and the LVQ algorithm). Finally, segment profiles for binary data can be depicted in a more informative way by visualizing bootstrap replications with box plot diagrams. The target audience for this paper thus consists of both academics and practitioners interested in explorative partitioning techniques. (author's abstract) / Series: Working Papers SFB "Adaptive Information Systems and Modelling in Economics and Management Science"

Zugl.: Mannheim, Univ., Diss., 2006

Thesis (doctoral)--Hochschule St. Gallen für Wirtschafts-, Rechts- und Sozialwissenschaften, 1999.

Zugl.: Göttingen, Universiẗat, Diss., 1994.

Zugl.: Mannheim, Universiẗat, Diss., 2006.

The generation of random variates from multivariate binary distributions has not gained as much interest in the literature as, e.g., multivariate normal or Poisson distributions. Binary variables are important in many types of applications. Our main interest is in the segmentation of marketing data, where data come from customer questionnaires with "yes/no" questions. Artificial data provide a valuable tool for the analysis of segmentation tools, because data with known structure can be constructed to mimic situations from the real world (Dolnicar et al. 1998). Questionnaire data can be highly correlated, when several questions covering the same field are likely to be answered similarly by a subject. In this paper we present a computationally fast method to simulate multivariate binary distributions with a given correlation structure. The implementation of the algorithm in R, an implementation of the S statistical language, is described in the appendix. / Series: Working Papers SFB "Adaptive Information Systems and Modelling in Economics and Management Science"

Zugl.: Münster (Westfalen), Universiẗat, Diss., 2009.

Zugl.: Münster (Westfalen), Univ., Diplomarbeit, 2002