Hybrid C&RT-Logit Models In Churn Analysis

• Authors: Mariusz Łapczyński
• Languages of publication: EN

Abstracts

EN

This article attempts to explain and predict the termination of relationships in telecommunications services by using the hybrid C&RT-logit model. The combination of decision trees (C&RT algorithm) with the logistic model enriches the model interpretation and sometimes improves the accuracy of prediction. Decision trees permit to detect interactions among variables and make the model resistant to outliers and to lack of data. On the other hand, the logistic model can extend the interpretation by using odds ratios. The solution delivered by the hybrid approach was compared with the decision tree model and the logistic model. Due to the difficulty in obtaining the real dataset from the Polish market, it was decided to build a model based on the data obtained from the repository http://www.dataminingconsultant.com/DMMM.htm . The models’ performance was estimated by using popular measures such as accuracy, recall, precision, true negative rate, G-mean, F measure and lift charts.

Keywords

EN

churn analysis; hybrid C&RT-logit model; C35; M31

• Publisher: Sciendo
• Journal: Folia Oeconomica Stetinensia
• Year: 2014
• Volume: 14
• Issue: 2
• Pages: 37-52

Dates

published

2014-12-01

received

2014-04-14

accepted

2014-10-24

online

2015-06-03

Contributors

author

Mariusz Łapczyński

• Cracow University of Economics, Department of Market Analysis and Marketing Research, Rakowicka 27, 31-510 Cracow, Poland

References

• Breiman, L., Friedman, J., Stone, C.J. & Olshen, R.A. (1984). Classification and Regression Trees. New York: Chapman and Hall.
• Buckinx, W. & Van den Poel, D. (2005). Customer base analysis: partial defection of behaviourally loyal clients in a non-contractual FMCG retail setting. European Journal of Operational Research, 164, 252–268.
• Burez, J. & Van den Poel, D. (2007). CRM at a pay-TV company: Using analytical models to reduce customer attrition by targeted marketing for subscription services. Expert Systems with Applications, 32, 277–288.
• Christopher, M., Payne, A. & Ballantyne, D. (2008). Relationship Marketing. Creating Stakeholder Value, 5 ed. Oxford: Butterworth-Heinemann.
• Coussement, K. & Van den Poel, D. (2008). Churn prediction in subscription services: An applications of support vector machines while comparing two parameter-selection techniques. Expert Systems with Applications, 34, 313–327.
• Customer Retention: A Business Imperative (2004). Baxter Strategies Incorporated. New York: Woodbury.
• Farris, P.W., Bendle, N.T., Pfeifer, P.E. & Reibstein, D.J. (2006). Marketing Metrics: 50+ Metrics Every Executive Should Master. New Jersey: Wharton School Publishing.
• Giudici, P. & Dequarti, E. (2011). Statistical Models to Predict Academic Churn Risk. In: Classification and Multivariate Analysis for Complex Data Structures, eds. B. Fichet, D. Piccolo, R. Verde, M. Vichi (pp. 41–49). Berlin, Heidelberg: Springer-Verlag.
• Gummesson, E. (2008). Total relationship marketing. 3 ed. Oxford: Butterworth-Heinemann.
• Gupta, S. & Zeithaml, V. (2006) Customer Metrics and Their Impact on Financial Performance. Marketing Science, 29 (6), 718–739.[Crossref]
• Idris, A., Khan, A. & Lee, Y.S. (2013). Intelligent churn prediction in telecom: employing mRMR feature selection and RotBoost based ensemble classification. Applied Intelligence, 39, 659–672.[Crossref][WoS]
• Jeffery, M. (2010). Data-Driven Marketing. The 15 Metrics Everyone in Marketing Should Know. Hoboken, NJ: John Wiley & Sons.
• Kim, S., Shin, K.S. & Park, K. (2005). An Application of SVM for Customer Churn Analysis: Credit Card Case. In: Advances in Natural Computation. Lecture Notes in Computer Science, eds. L. Wang, K. Chen, Y.S. Ong (pp. 636–647). Berlin, Heidelberg: Springer-Verlag. Vol. 3611.
• Liao, K.H. & Chueh, H.E. (2011). Applying Fuzzy Data Mining to Telecom Churn Management. In: Intelligent Computing and Information Science. Communications in Computer and Information Science, ed. R. Chen (pp. 259–264). Berlin, Heidelberg: Springer-Verlag. Vol. 134.
• Lindahl, W.E. & Winship, C. (1994). A Logit Model with Interactions for Predicting Major Gift Donors. Research in Higher Education, 35 (6), 729–743.[Crossref]
• Łapczyński, M. (in press), The use of hybrid predictive C&RT-logit models in analytical CRM, Second European Conference on Data Analysis, ECDA 2014, July 2–4, Jacobs University, Bremen, “Program and Abstracts”, .
• Miguéis, V.L., Van den Poel, D., Camanho, A.S. & Falcão e Cunha, J. (2012). Predicting partial customer churn using Markov for discrimination for modeling first purchase sequences. Advances in Data Analysis and Classification, 6, 337–353.
• Morik, M. & Köpcke, H. (2004). Analysing Customer Churn in Insurance Data – A Case Study. In: Proceedings of the 8th European Conference on Principles and Practice of Knowledge Discovery in Databases, eds. J.F. Boulicaut, F. Esposito, F. Giannotti, D. Pedreschi (pp. 325–336). Berlin, Heidelberg: Springer-Verlag.
• Naveen, N., Ravi, V. & Rao, C.R. (2010). Data Mining via Rules Extracted from GMDH: An Application to Predict Churn in Bank Credit Cards. In: Knowledge-Based and Intelligent Information and Engineering Systems. Lecture Notes in Computer Science, eds. R. Setchi, I. Jordanov, R.J. Howlett, LC. Jain (pp. 80–89). Berlin, Heidelberg: Springer-Verlag. Vol. 6276.
• Phadke, C., Uzunalioglu, H., Mendiratta, V.B., Kushnir, D. & Doran, D. (2013). Prediction of Subscriber Churn Using Social Network Analysis. Bell Labs Technical Journal, 17 (4), 63–76.[WoS][Crossref]
• Sheth, J.N. & Parvatiyar, A. (2000). Relationship Marketing in Consumer Markets. Antecedents and Consequences. In: Handbook of Relationship Marketing, eds. J.N. Sheth, A. Parvatiyar (pp. 171–208). Thousand Oaks, CA: Sage Publications.
• Steinberg, D. & Cardell, N.S. (1998): The hybrid CART-logit model in classification and data mining, (14.10.2007).
• Van den Poel, D. & Larivière, B. (2004). Customer attrition analysis for financial services using proportional hazard models. European Journal of Operational Research, 157, 196–217.

Identifiers

DOI

10.1515/foli-2015-0006