|
Activity Number:
|
125
- Practical Recommendations for Prediction Modeling That Advance Innovation
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, August 8, 2022 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
|
Abstract #319229
|
|
|
Title:
|
Current Methods for Evaluating Prediction Model Performance
|
|
Author(s):
|
Michael W. Kattan*
|
|
Companies:
|
Cleveland Clinic
|
|
Keywords:
|
prediction models
|
|
Abstract:
|
Innovative published prediction models vary tremendously regarding the way their performance is assessed. Numerous metrics and plots have been introduced with little overall guidance for choosing which to report. This problem becomes particularly critical when rival prediction models are being compared, as one metric may favor one model while another metric favors another. This presentation will discuss some straightforward solutions to prediction model assessment and comparison.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2022 program
|
