I read some papers of risk prediction models, and generally there are three types of measures of the performance: a) Total model performance (e.g. R^2); b) Discrimination; c) Calibration.
However, many papers of risk prediction model only reported the Discrimination and Calibration, seldom reported the R^2. What confused me is that if R^2 is important in risk prediction models. And is there a threshold for the minimal value of R^2 in such models?
Chang Xu -
I am not familiar with risk prediction models (some kind of logistic regression???), but I think that in general, R-squared is not very helpful. Please note the following, for example, where they note that R-squared is not a measure of fit: https://data.library.virginia.edu/is-r-squared-useless/
I think that your model may still be good, but the variance may be high with large R-squared.
Perhaps, as in many cases, you may learn more about your application by examining various scatterplots.
Cheers - Jim
- Badges
- Science topic
- Similar topics
- Counseling Psychology
- Confusion