Please implement simultaneous prediction intervals in the distribution platform. This type of interval sits in between a prediction interval that predicts where a single future value will fall and a tolerance interval that predicts where all future values will fall. The simultaneous prediction interval allows to set the interval for a certain number of future values.
See a description here: https://search.r-project.org/CRAN/refmans/EnvStats/html/predIntNormSimultaneous.html
What inspired this wish list request? Acceptance criteria for pharmaceutical comparability studies must be set. For certain quality attributes, it is appropriate to define a range based on a number of pre-change manufactures batches, e.g., six. Post-change, you will usually just have three batches. Currently, the best way is to calculate a tolerance interval based on the pre-change batches, and then check if the 3 post-change batches fall within in. With the simultaneous prediction interval it would be possible to set a range that is precisely expected for a given number of post-change batches.
What is the improvement you would like to see? The Distribution platform could have this option next to confidence and tolerance intervals.
Why is this idea important? Enable appropriate setting of acceptance criteria for pharmaceutical comparability studies.