Off-campus South Dakota State University users: To download campus access theses, please use the following link to log into our proxy server with your South Dakota State University ID and password.
Non-South Dakota State University users: Please talk to your librarian about requesting this thesis through interlibrary loan.
Thesis - University Access Only
Master of Science (MS)
Mathematics and Statistics
Credit scoring has evolved into a critical tool for assessing risk in consumer lending. This thesis analyzes the standard process of consumer credit scoring including variable selection, coarse classification, model building, and model evaluation. At each stage, popular techniques and the math explaining them are examined. An in depth discussion of the origin of the information value and how the weights‐of‐evidence transformation is related to logistic regression is presented. In addition, a revolutionizing way credit reporting agencies will soon store their data and a new technique to leverage this new data will be introduced. In this technique, velocity variables are created and added to the standard logistic regression model to account for the trend in behavior of credit card consumers. Concluding, a case study comparing the standard consumer credit scoring method to one that leverages the velocity variables available with the new credit bureau data to predict consumer attrition is conducted. Various fit statistics comparing the methods indicated that adding the velocity variables provided significant lift. This new technique is one of many likely to be implemented by modelers to leverage the new data soon available from credit bureaus reporting agencies.
Library of Congress Subject Headings
Credit scoring systems
Includes bibliographical references (pages 58-59)
Number of Pages
South Dakota State University
In Copyright - Non-Commercial Use Permitted
Burton, Ryan, "Consumer Credit Scoring and an Enhanced Data Mining Technique" (2013). Electronic Theses and Dissertations. 2084.