| Title | Churn and Net Promoter Score forecasting for business decision-making through a new stepwise regression methodology |
|---|---|
| Authors | Velez, D. , Ayuso, A. , PERALES GONZÁLEZ, CARLOS, Tinguaro Rodriguez, J. |
| External publication | No |
| Means | Knowl Based Syst |
| Scope | Article |
| Nature | Científica |
| JCR Quartile | 1 |
| SJR Quartile | 1 |
| JCR Impact | 8.038 |
| SJR Impact | 1.587 |
| Web | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85082480132&doi=10.1016%2fj.knosys.2020.105762&partnerID=40&md5=c37e849fb057f6c8df7f470564ff59a7 |
| Publication date | 21/05/2020 |
| ISI | 000527301700010 |
| Scopus Id | 2-s2.0-85082480132 |
| DOI | 10.1016/j.knosys.2020.105762 |
| Abstract | Companies typically have to make relevant decisions regarding their\n clients\' fidelity and retention on the basis of analytical models\n developed to predict both their churn probability and Net Promoter Score\n (NPS). Although the predictive capability of these models is important,\n interpretability is a crucial factor to look for as well, because the\n decisions to be made from their results have to be properly justified.\n In this paper, a novel methodology to develop analytical models\n balancing predictive performance and interpretability is proposed, with\n the aim of enabling a better decision-making. It proceeds by fitting\n logistic regression models through a modified stepwise variable\n selection procedure, which automatically selects input variables while\n keeping their business logic, previously validated by an expert. In\n synergy with this procedure, a new method for transforming independent\n variables in order to better deal with ordinal targets and avoiding some\n logistic regression issues with outliers and missing data is also\n proposed. The combination of these two proposals with some competitive\n machine-learning methods earned the leading position in the NPS\n forecasting task of an international university talent challenge posed\n by a well-known global bank. The application of the proposed methodology\n and the results it obtained at this challenge are described as a\n case-study. (C) 2020 Elsevier B.V. All rights reserved. |
| Keywords | Churn prediction; Net Promoter Score; Stepwise regression; WOE variables |
| Universidad Loyola members |
|