Hi Ruby,
Thanks for getting back to me.
1. Using a manual annotation method I described should work for you.
2. JMP will not be able to calculate the AUC in graph builder. It does do this for Logistic Regression, some curve fits for Non-linear platform, Partition and some other categorical response based modeling methods. For the comparison, if you goal is to say there is a statistical difference, a t-test or analysis of means would be good enough and you could get the mean difference between the groups. But if you are trying to quantitate the amount of difference, then I can see why you would want to use AUC. Maybe summing up the means at across the time points for each of the two groups and then taking the difference would give you an idea of the difference, You can used a combination of Tabulate (time as rows, glucose as a column with condition nested within), then make into data table, and then make a column formula to make the difference of each time point and/or the sum of the columns . Somebody else might have a better idea.
As for the bar chart you are trying to recreate, unfortunately the only way to change the fill color and still have a black boarder may be restricted to scripting it at this point. One can change the colors in to a Black and White scale using the color drop zone in graph builder and then changing each color of the groups one at a time or by using a color scheme. Now if you used an outlier box plot that Jim shows, then his method would work quite well.
Chris Kirchberg, M.S.2
Data Scientist, Life Sciences - Global Technical Enablement
JMP Statistical Discovery, LLC. - Denver, CO
Tel: +1-919-531-9927 ▪ Mobile: +1-303-378-7419 ▪ E-mail: chris.kirchberg@jmp.com
www.jmp.com