Closed JanionNevill closed 2 months ago
I've found two problems. In the certolizumab data set (attached)
- one of the colours doesn't match between the line and the region.
- if you add all treatments, add all the confidence regions, then remove them one by one, at some point it fails.
Those 2 issues turned out to have the same root cause. I fixed this by adding NA
s for intervals where a region was present, and vice versa.
Added background task to calculate confidence regions.