0

I have a data frame with many factor variables of multiple level and an outcome variable. I wanted to calculate the sensitivity and specificity the factor variables against the outcome variable with 95% CI and the area under ROC curve for each variable . Is there any method to do this in rstudio?

Berhe
  • 1
  • 2
  • 1
    You need binary predictions from a model to calculate the AUROC. Where are they? – danlooo Mar 17 '22 at 15:44
  • I have no idea how fever can be binary if you want to predict it, but here is a tutorial on binary classification and performance evaluation including AUROC: https://www.tidymodels.org/start/case-study/ – danlooo Mar 18 '22 at 15:12
  • We still need a model here for prediction. Should all COVID patients be predicted as having fever? Don't make much sense. So we need other predictors here (other many factors as written in the question). You first need to train a model (e.g. random forest). This will tell you how to combine the many factors including covid to predict fever – danlooo Mar 18 '22 at 15:52

0 Answers0