ROC analysis

#1
Hi, Im not sure if this is the right place, if not please move it to the correct forum.

I have an ROC curve generated using medcalc (400 observations, 100 positive and 300 negative) and what I would like to know is if there is any way I can find out what the sensitivity and specificity of a specific cut point (not in the original sample) might be using the data for that curve, i.e. say in my sample the most accurate cut off point was 789 and I wanted to know what the sensitivity and specificity would be for a cut point of 734 how would I do this?

Thanks
 

Karabiner

TS Contributor
#3
It is not easy to understand where your problem lies. If you know how
sensitivity an specifity are calculated, you can easily construct
2x2 tables for predicted outcome versus actual outcome. And for
determining predicted outcome you can use your cutoff point(s)
of interest.

Regards

K.