How can determine the best cut point with large number of observation?

[Haytham Shebel]

I have data set contains 300 observation and when I am using the Roctab command to calculate sensitivity. and specificity the result is a long table of 300 observation each has its sensitivity, specificity and cut-point.

This an example of the code I have used used [roctab sex rocl, graph summary detail].

Is there any way or code to find out the best cut-point without searching this long table? Also, How can I calculate PPV, NPV and total accuracy using Stata.

I am using Stata Version 15.1.
Thanks for your expected help.

[Clyde Schechter]

What do you mean by "best" cut-point? What criterion would you apply to the combination of sensitivity and specificity at each cutoff to evaluate which ones are better and which ones worse. If you figure out what criterion you want, coding it is usually quite simple.

From a decision theoretic perspective, it is not possible to choose a best cut-off based only on sensitivity and specificity. Rather, the best cutoff would have to take into account the consequences of false positives and false negatives, as well as the actual proportion of positive cases in the target population.