A Combination of Multiperiod Training Data and Ensemble Methods in Churn Classification: the Case of Housing Loan Churn

Customer retention has been the focus of customer relationship management research in the financial sector during the past decade. The first step in customer retention is to classify the customers into binary groups of possible churners, meaning customers that are likely to switch to another service provider, and non-churners, referring to those that are probably staying with the current provider. The second step in customer retention is to take action to retain the most probable churners to either minimize costs or maximize benefits. As a result, churn classification is an important first step in customer retention. However, the main challenge in churn classification is the extreme rarity of churn events. For example, the churn rate in the banking industry is usually less than 1%. In order to overcome this rarity issue, a great deal of research has been found to improve the two main aspects of a churn classification model: the training data and the algorithm. Regarding the training data, the recently proposed multi-period training data approach is found to outperform the single period training data thanks to the more effective use of longitudinal data of churn behavior. Regarding the churn classification algorithms, the most advanced and widely employed is ensemble method, which combines multiple models to produce a more powerful one. Two popularly used ensemble techniques are random forest and gradient boosting, both of which are found to outperform logistic regression and decision tree in classifying churners from non-churners. To the best of the author’s knowledge, the proposed multi-period training data has not been applied to the ensemble methods in a churn classification model. As a result, the thesis would like to study whether this multi-period training data approach, when employed together with ensemble methods in a churn classification model, produces better churn prediction than with logistic regression and decision tree. The ensemble methods used in thi