Market segmentation via mixtures of constrained factor analyzers
2013
Online
Elektronische Ressource
In this paper we introduce a procedure for the parameter estimation of mixtures of factor analyzers, which maximizes the likelihood function in a con- strained parameter space, to overcome the well known issue of singularities and to reduce spurious maxima of the likelihood function. A Monte Carlo study of the per- formance of the algorithm is provided. Finally the proposed approach is employed to provide a market segmentation, to model a set of quantitative variables provided by a telecom company, and related to the amount of services used by customers
Titel: |
Market segmentation via mixtures of constrained factor analyzers
|
---|---|
Link: | |
Veröffentlichung: | 2013 |
Medientyp: | Elektronische Ressource |
Schlagwort: |
|
Sonstiges: |
|