Seminars are held weekly on Wednesdays, from 3-4pm central time in Hardin NW 49. Please attend in person if possible.
If you need a zoom link, please fill out the form at the bottom of this page to receive an invitation.
January 19 - Charlie Beer
Title: A Custom Application of GLM and Weight of Evidence Binning to Assign Credit Default Ratings
Abstract: Assessments of default risk are among the most important aspects of risk management for a lending institution. We will present an application of generalized linear modeling on variables binned via weight of evidence procedures to provide a custom default risk score – analogous to a credit score. While this method is similar in many respects to a standard scorecard, several differences in approach and methodology make for a more robust and flexible model. This approach allows for the control of time-dependent influences and interactions between the predictive variables. It also has advantages in variable selection. The final presentation of the model maintains the simplicity of a standard scorecard while improving predictive performance. The approach/methodology and model results will both be discussed.
Seminar Zoom Link Request
(if you are a Stats Grad Student, the Zoom link can be found in your calendar invite)