CODEWITHSUNDEEP
rrocauc
striatum
striatum

Reputation: 1598

Testing the difference between two AUCs

Testing the difference between two AUCs seems to be straightforward. There are some relevant sources, like this one and links therein.

I have an extremely simple case of two ROCs, similar to this example:

simple plot.

It is quite easy to calculate the AUCs using the Trapezoidal Rule. However, the tests of difference assume that the variance is calculated, and I cannot wrap my head around how to do that. Variance of what? The data I have is simply something like this (data is, of course, aggregated):

Hits  FalseAlarms
 0.0          0.0
 0.9          0.3
 1.0          1.0

How can this be worked out in R using the input data like above? Thanks.

Upvotes: 0

Views: 12

Answers (0)

Related Questions