Hi everyone,
When I tried to evaluate how different phenotype algorithms ¶ perform in the evaluation cohort, I found that the sum of true positives (case & belonging to PA) and false positives (non-case & belonging to PA) is significantly lower than PA counts. I tried to adjust parameters such as baseSampleSize to a larger number in order to factor in a larger evaluation cohort. But it didn’t change the result. How can I fix this problem? Thanks!
