Estimate significance of difference in two AUCs by resampling.
testAUCpair( d, model1Name, model2Name, yName, yTarget = TRUE, ..., na.rm = FALSE, returnScores = FALSE, nrep = 100, parallelCluster = NULL )
d | data.frame |
---|---|
model1Name | character model 1 column name |
model2Name | character model 2 column name |
yName | character outcome column name |
yTarget | target to match to y |
... | extra arguments (not used) |
na.rm | logical, if TRUE remove NA values |
returnScores | logical if TRUE return detailed resampledScores |
nrep | number of re-sample repetition to estimate p value. |
parallelCluster | (optional) a cluster object created by package parallel or package snow |
AUC pair test
set.seed(25325) d <- data.frame(x1=c(1,2,3,4,5,6,7,7), x2=1, y=c(FALSE,TRUE,FALSE,FALSE, TRUE,TRUE,FALSE,TRUE)) testAUCpair(d,'x1','x2','y',TRUE)#> [1] "AUC test resampled AUC1>AUC2: (AUCs=0.6562;0.5, s.d.=0.2033, e=n.s.)."