Statistical Methods In Customer Relationship Management PDF Download

This dissertation, "Stochastic Models for Customer Relationship Management" by Ka-kuen, Wong, 黃嘉權, was obtained from The University of Hong Kong (Pokfulam, Hong Kong) and is being sold pursuant to Creative Commons: Attribution 3.0 Hong Kong License. The content of this dissertation has not been altered in any way. We have altered the formatting in order to facilitate the ease of printing and reading of the dissertation. All rights not granted by the above license are retained by the author. Abstract: Abstract of thesis entitled STOCHASTIC MODELS FOR CUSTOMER RELATIONSHIP MANAGEMENT submitted by WONG Ka-Kuen for the degree of Master of Philosophy at The University of Hong Kong in July 2004 Customer relationship management is the development and maintenance of a long-term relationship with the customers. It is a crucial factor for surviving in a competitive market. Classication of customers and promotion budget allocation arethecoreissuesinrelationshipmarketing. Deterministicmodelshavebeenused in customer relationship management. In practice, however, most of them are not suitable because of the stochastic nature of the problem. In this thesis, a series of stochastic models and numerical algorithms are proposed and evaluated. Classication of customers plays an important role in promotion planning and budget setting. Apart from many available tools for classication, one poten- tial tool is the Hidden Markov Models (HMMs) and it can be implemented in a Microsoft Excel worksheet. Ecient parameter estimation methods with fast numerical algorithms for HMMs are also proposed. Numerical examples are given to demonstrate the model's e(R)ectiveness and eciency of the estimation method. Promotionbudgetallocationisdesignedtoattractandkeepcustomers. Stochas- tic dynamic programming models with the Markov chain techniques are proposed for capturing the customer behaviour. The model can be implemented easily and eciently in a Microsoft Excel worksheet, and precise implementation is also demonstrated and discussed in details. Furthermore, a higher-order Markov chain modelisafurtherwayofcapturingthedynamicsofpracticaldata. Ahigher-order Markov decision process is proposed. Practical data of a computer servicecompanyisusedtoillustratethee(R)ectivenessoftheproposedhigher-orderMarkov model. Aremarkableimprovementinpredictingthecustomerbehaviorisobserved. DOI: 10.5353/th_b3028996 Subjects: Customer relations - Management - Statistical methods Stochastic processes