지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학술대회자료
- 저자정보
- 발행연도
- 2020.8
- 수록면
- 106 - 127 (22page)
이용수
초록· 키워드
오류제보하기
Purpose: Human activities tend to burst at specific times, followed by long dormant periods. This study aims to show whether an entropy measure of non-Poisson trading patterns has advantages over the canonical recency, frequency, and monetary value framework in customer churn prediction.
Design/methodology/approach: This study analyzes detailed trading records, as well as the demographic profiles of 486,049 customers from a major securities company in Korea and explore the extent to which a measure of clustered purchases is linked to customers’ future dormant status. Motivated by the recent development of statistical learning techniques, I employ three different classification methods: (a) logistic regression, (b) a gradient boosting classification tree, and (c) a neural network.
Findings: The LASSO logistic regression, the information gain metric in gradient boosting decision trees, and the relative importance method in neural networks all lend support to the conclusion that the clumpiness measure of trade clustering plays a significant role in explaining customers’ future churning. Furthermore, recently developed machine learning techniques reduces churn prediction errors to a greater extent.
Practical implications: The top decile lift is at least five times as large as the sample attrition rate, meaning that the density of churned customers in the top decile is five times the density of churned customers in the sample, at the very least. The magnitude of the observed top decile lift demonstrates that the estimated model can be effectively used for targeting a customer group in retention campaigns.
Originality/value: Traditional survey-based perception-like metrics such as brand recognition, satisfaction, trust, and loyalty target at small and selected groups of customers, and thus, do not capture the true value of customers in the population. On the contrary, this study analyzes an extensive data set using up-to-date quantitative techniques and show that a metric-based parsimonious RFMC approach coupled with machine learning techniques can be effectively used to better gauge customer lifetime value.
Design/methodology/approach: This study analyzes detailed trading records, as well as the demographic profiles of 486,049 customers from a major securities company in Korea and explore the extent to which a measure of clustered purchases is linked to customers’ future dormant status. Motivated by the recent development of statistical learning techniques, I employ three different classification methods: (a) logistic regression, (b) a gradient boosting classification tree, and (c) a neural network.
Findings: The LASSO logistic regression, the information gain metric in gradient boosting decision trees, and the relative importance method in neural networks all lend support to the conclusion that the clumpiness measure of trade clustering plays a significant role in explaining customers’ future churning. Furthermore, recently developed machine learning techniques reduces churn prediction errors to a greater extent.
Practical implications: The top decile lift is at least five times as large as the sample attrition rate, meaning that the density of churned customers in the top decile is five times the density of churned customers in the sample, at the very least. The magnitude of the observed top decile lift demonstrates that the estimated model can be effectively used for targeting a customer group in retention campaigns.
Originality/value: Traditional survey-based perception-like metrics such as brand recognition, satisfaction, trust, and loyalty target at small and selected groups of customers, and thus, do not capture the true value of customers in the population. On the contrary, this study analyzes an extensive data set using up-to-date quantitative techniques and show that a metric-based parsimonious RFMC approach coupled with machine learning techniques can be effectively used to better gauge customer lifetime value.
목차
Abstract
1. Introduction
2. Quantifying clustered purchases
3. Research design
4. Data
5. Results
6. Discussion and implications
7. Limitations and Future Research
8. Conclusion
References
참고문헌 (0)
참고문헌 신청함께 읽어보면 좋을 논문
논문 유사도에 따라 DBpia 가 추천하는 논문입니다. 함께 보면 좋을 연관 논문을 확인해보세요!
-
CUSTOMER FIRST
한국서비스경영학회 학술대회
2022 .05
-
만족-충성 관계에 영향을 미치는 고객 네트워크의 특징
서비스경영학회지
2020 .09
-
서비스 품질이 고객만족에 미치는 영향 : 고객 간 상호작용과 변화주도행위의 조절된 매개효과
대한경영학회지
2019 .01
-
A Study on the Antecedent and Consequences of Customer Value : Focused on Overseas Tourism
인터넷전자상거래연구
2018 .10
-
이동통신 이용자의 지각된 고객가치가 고객만족도 및 고객충성도에 미치는 영향: 고객자산가치 모형을 기반으로
인터넷전자상거래연구
2019 .06
-
고객서비스 정보시스템 재구축과 신규구축 성공에 영향을 미치는 요인에 관한 비교사례연구
지식경영연구
2019 .01
-
치과의료서비스에서 고객지향성이 관계의 질과 충성도에 미치는 영향에 관한 연구
의료경영학연구
2017 .01
-
글로벌 스포츠브랜드 기업의 고객지향성이 중국소비자의 충성도에 미치는 영향에 관한 연구
유통경영학회지
2018 .06
-
토픽 분석을 활용한 관심 기반 고객 세분화 방법론
Journal of Information Technology Applications & Management
2015 .03
-
호텔레스토랑 고객가치가 고객희생과 고객충성도의 관계에 미치는 매개효과
Tourism Research
2017 .01
-
고객지원인식, 고객-종업원 교환관계, 고객행복 및 고객시민행동 간의 관계
마케팅논집
2017 .01
-
Customer Experience and Repurchase Intention in Multi-Channel: Customer Satisfaction as Mediating Variable
산경연구논집
2021 .01
-
호텔레스토랑 고객가치가 고객만족과 고객충성도에 미치는 영향-고객만족의 매개효과와 전환비용의 조절효과-
Tourism Research
2017 .01
-
서비스 접점에서 종업원의 행동이 고객협력과 공동생산에 미치는 영향
대한경영학회지
2018 .09
-
The Effect of Product Quality and SERVQUAL on Customer Satisfaction and Customer Loyalty
전문경영인연구
2018 .01
-
Relationships between Customer Socialization, Customer Participation, and Loyalty of On-line Service Providers
유통과학연구
2015 .01
-
Negative Spillover Effects of Other-Customer Failure in Airline Context
유통과학연구
2017 .01
-
증강현실 기반의 제품체험으로 인한 고객 인게이지먼트 효과 연구 - 고객가치 공동창출의 매개효과 분석 -
소비자학연구
2018 .01
이 논문의 저자 정보
최근 본 자료
댓글(0)
0
UCI(KEPA) : I410-ECN-0101-2020-324-001158439