I have a sample of cancer patients, of whom 41% died due to the cancer and only 9% died from competing events. So the number of deaths due competing events is fewer than the number of deaths due to cancer. Is it possible to use techniques used to deal with class imbalance in the Fine and Gray competing risk model? If yes, which technique would be most suitable for this kind of a problem?
Asked
Active
Viewed 44 times
2
-
3See Are unbalanced datasets problematic, and (how) does oversampling (purport to) help? – user2974951 Mar 29 '23 at 06:50