Andrew Vickers
@vickersbiostats
Biostatistician at Memorial Sloan Kettering Cancer Center. Special interest in prostate cancer, risk prediction, patient-reported outcomes, decision-making.
ID: 1225800762447532034
https://www.mskcc.org/profile/andrew-vickers 07-02-2020 15:17:42
2,2K Tweet
5,5K Followers
73 Following
Venk Murthy MD PhD You are aware that these are completely wrong? It depends on whether the predictor is continuous or binary and, if the latter, the prevalence of the predictor. If a marker is found in 0.1% of population, has an OR of 20, AUC will be ~0.505.