Hi. I wrote yesterday about evaluating a desirability function on new data and that solution is good, assuming I want to use the predicted values. Now I have another question.
For example, suppose I start with a dataframe that has 4 feature columns, 3 response columnns and a grouping column. I fit the 3 responses by the group on the 4 features (the doe is RSM so this all works well). Now I have 2 groups of least squares fits in my Fit Model and a profiler for each. I somehow, in another investigation, was able to add a column to my dataframe that has the Desirability Formula in it --- so the title of the column is Desirability and it evaluates the desiriability on the actual measured parameters -- but I can't seem to duplicate this behavior, even if I separate the data frame into two so that I am only fitting on one of the grouping variables. I thought the right thing to do was in the prediction profiler on the fit model page select Optimization and Desirability -> Save Desirability Formula. But nothing happens and I have no idea why not and why it did work once before. Any thoughts/advice? Thanks!
