Katalina Aganesan
My feedback
4 results found
-
1 vote
-
2 votes
-
26 votes
An error occurred while saving the comment An error occurred while saving the comment Assessing credit scoring models assessment becomes even more critical when alternative data is involved. The article on RiskSeal offers a solid overview, but I believe deeper metrics beyond AUC—like stability over time and fairness analysis—are essential. Has anyone tried evaluating model drift when using non-traditional data sources?
-
1 vote
Thanks for sharing this! I completely agree—accuracy alone isn't enough when evaluating modern credit scoring models. Fairness, transparency, and compliance are just as critical, especially with the rise of alternative data. I found the article you mentioned really helpful—it breaks down the challenges and gives a solid framework for assessing effectiveness in real-world scenarios.