顧客價值分析隨機模型建立之研究
Other Title
Stochastic Modeling of Customer Value Analysis
Date Issued
2005
Date
2005
Author(s)
DOI
932416H002013
Abstract
With the rapid progress of information technology
and development of related algorithms, customers’
lifetime value can now be better estimated. The goal of
this research is to propose a systematic method to estimate
customers’ lifetime value and conduct empirical
validation.
Our proposed model consists of the following steps:
1. Construct a stochastic model of customers’ purchase
behavior and propose the underlying probability
distributions of customers’ purchase frequency and
monetary.
2. Model the customer’ purchase behavior as a Markov
chain and the associated transition states based on
RFM (recency frequency monetary) model.
3. Use Bayesian theorem to derive transition state
probability.
4. Construct customers’ transition and reward matrices
and calculate customers’ lifetime value.
The proposed method has been validated using an
industrial computer company’s sales data. Results show
that:
1. Our model outperforms the current empirical
estimated practices and achieves a very accurate
estimation of customers’ value.
2. The proposed method is able to model customers’
heterogeneity and predict each customer’s purchase
behavior.
Subjects
Customer value analysis
Markov chain
Bayesian statistics
Publisher
臺北市：國立臺灣大學工商管理學系
Type
report
File(s)![Thumbnail Image]()
Loading...
Name
932416H002013.pdf
Size
150.65 KB
Format
Adobe PDF
Checksum
(MD5):7c450796c9e7ec45c661096fb798da06