Optimizing Decision Making for Retail Lending on the Basis of a Customer Profitability Model

This present Thesis is aimed at solving the actual problem of finding the optimal method for decision making in retail lending. The first chapter describes the theoretical background of the study: explains the classical approach to this problem, identifies its shortcomings and prerequisites for applying an approach based on customer profitability. The basic definitions and concepts are given; a review and analysis of the literature is done. The second chapter describes modern tools for developing decision-making models in retail lending. Some models based on customer profitability using linear regression and gradient boosting methods are built and analyzed by comparative analysis. The results of the study are of interest to specialists in the field of credit scoring and finance, dealing with issues of maximizing bank profits, and also to CRM specialists interested in attracting more profitable customers, and other banking analysts.