Timeline for Ignore strata in external validation of stratified Cox prop hazards model?
Current License: CC BY-SA 3.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| S Oct 23, 2014 at 23:16 | history | bounty ended | purple51 | ||
| S Oct 23, 2014 at 23:16 | history | notice removed | purple51 | ||
| Oct 23, 2014 at 23:16 | vote | accept | purple51 | ||
| Oct 23, 2014 at 13:13 | answer | added | Frank Harrell | timeline score: 1 | |
| Oct 23, 2014 at 12:41 | history | edited | purple51 | CC BY-SA 3.0 |
Gave example
|
| Oct 20, 2014 at 5:01 | comment | added | purple51 | The main issue is the validation: I can validate C-index within each stratum separately, or I can evaluate over all individuals ignoring the strata, which leads to different results. | |
| Oct 16, 2014 at 12:21 | comment | added | Peter | You should still fit a stratified model. The model in training and validation part should remain the same. | |
| S Oct 16, 2014 at 11:48 | history | bounty started | purple51 | ||
| S Oct 16, 2014 at 11:48 | history | notice added | purple51 | Draw attention | |
| Oct 14, 2014 at 10:20 | history | asked | purple51 | CC BY-SA 3.0 |