search

Thie1e / cutpointr

Optimal cutpoints in R: determining and validating optimal cutpoints in binary classification
https://cran.r-project.org/package=cutpointr
85 stars 13 forks source link

An ambiguous region bounded by two cutpoint #62

Open jwang-lilly opened 1 year ago

jwang-lilly commented 1 year ago

Hi Christian:

When developing a diagnostic test, it is common to have three regions: eg high confident negative, ambiguous, and high confident positive. Two cutpoints cp1 and cp2 would accomplish this. Cutpointr current only accepts one cp as: method = oc_manual, cutpoint = cp. I ended up preparing two data, data1 includes only the ambiguous middle and data2 without those falling in the middle region.

It would be great if cutpointr could support something like:

opt_cut <- cutpointr(data=suicide, x=age, class=suicide, direction = ">=", pos_class = "yes", neg_class = "no", method = oc_manual, cutpoint = c(30, 40))

PrionMike commented 1 month ago

Hi Christian and jwang-lilly. I too am very interested in this question. Recognizing that it would require considerable developmental work, it seems to me that cutpointr would be even better if it could help tackle this common problem. Some good progress has already been made to develop R functionality -- for an additional example, see https://www.mdpi.com/2075-4418/8/2/32.