Description Usage Arguments Value References Examples
This function can be used to calculate and compare AUC estimates. If a single set of test measurements is provided, the corresponding AUC and variance values are computed. If two sets of test measurements are provided, a significance test for a difference in AUC estimates is also computed. Optionally, the previous can also be computed in the presence of clustered data per Obuchowski (1997). This function implicity assumes that higher values of test_1 and test_2 are correlated with the positive gold standard (status = TRUE)
1 |
test_1 |
the set of test measurments for which the AUC should be computed |
test_2 |
a second set of test measurments for which the AUC is computed and compared to the first set of test measurements |
status |
a logical vector representing the gold standard negative/positive status of an observaiton |
cluster |
variable identifying clusters |
a list containing
auc |
a vector of AUC estimates |
var |
the individual variance estimate or covariance matrix |
test_stat |
the test statistic for comparing the AUC estimates, distributed standard normal |
p_value |
the p-value associated with the AUC test comparison (2-sided) |
Obuchowski N.A. (1997) Nonparametric Analysis of clustered ROC Curve Data. Biometrics, 53, 567-578.
DeLong E.R., DeLong D.M, Clarke-Pearson D.L. (1988) Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics, 44, 837-845.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 | library(fastAUC)
n_obs <- 10000
n_clust <- 100
cluster <- sample(x = 1:n_clust, size = n_obs, replace = T)
status <- as.logical(rbinom(n = n_obs, size = 1, prob = 0.3))
test_1 <- ifelse(status,
rnorm(n = n_obs, mean = 15, sd = 20),
rnorm(n = n_obs,mean = 10, sd = 20))
test_2 <- ifelse(status,
rnorm(n = n_obs, mean = 20, sd = 20),
rnorm(n = n_obs,mean = 10, sd = 20))
auc(test_1 = test_1, status = status)
auc(test_1 = test_1, test_2 = test_2, status = status, cluster = cluster)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.