Skip to main content

This content is closed to future replies and is no longer being maintained or updated.
Links may no longer function. If you have a similar request, please write a new post.

Question

 Tetiana Gural (TetianaGural)
PEGA
Consulting Manager
Pegasystems Inc.
PL
TetianaGural Member since 2014 1 post
PEGA
Posted: Feb 23, 2017
Last activity: Mar 6, 2017
Posted: 23 Feb 2017 4:33 EST
Last activity: 6 Mar 2017 2:25 EST
Closed
Solved

What is the difference between model's performance and model's coefficient of concordance?

According to studying guides for Senior Pega Marketing Consultant 7.21, both seem to be representing how good the model is in differentiating between positive and negative behavior.

To see attachments, please log in.
Decision Management

Accepted Solution

Posted: 7 years ago
Posted: 24 Feb 2017 3:17 EST
Otto Perdeck (Otto_Perdeck)
PEGA
Director, Data Science, Machine Learning & AI
Pegasystems Inc.
NL

Hello Tetiana,

Great to hear from you again! We use "Coefficient of Concordance" as a measure of the model performance - really a way to express how well a model discriminates between "positives" and "negatives". Many different model performance measures exist (e.g. accuracy, KS etc etc) but we use CoC, which really is just an alias for what is commonly referred to as "AUC" (Area Under the Curve). In future releases we will start using AUC instead of CoC.

Whatever the name - this is a performance measure that deals well with imbalanced binary outcomes (e.g. many more "offered" than "accepted"), which is common for the use cases of our customers.

Cheers

Otto

PS there is a Mesh page that explains the CoC in detail: https://mesh.pega.com/docs/DOC-86581

 

To see attachments, please log in.

We'd prefer it if you saw us at our best.

Pega Collaboration Center has detected you are using a browser which may prevent you from experiencing the site as intended. To improve your experience, please update your browser.

Close Deprecation Notice