Multiple comparisons adjustments added to the Red Triangle for CategoricalRequest to have multiple comparisons adjustments added to the Red Triangle for Categorical. Essentially including functionality similar to what is buried in one of the many red triangles in Fit Model under “LSMeans Student’s t, LSMeans Tukey HSD, LSMeans Dunnett.” I think what is currently implemented in Categorical is similar to LSMeans Student’s t.
These calculations are already available in Fit Model.
Statistics and AnalyticsRe: Multiple comparisons adjustments added to the Red Triangle for Categorical - Status changed to: In DevelopmentWe are working on this for JMP 16. We'll use a new command, FDR Adjusted PValues, which will compute False-Discovery-Rate-adjusted PValues (Benjamini and Hochberg, 1995).
Re: Multiple comparisons adjustments added to the Red Triangle for CategoricalHello. Just wanted to check in to see if there has been any progress on the need for multiplicity adjustments within Categorical platform (structured). Ideally a red triangle option. Repeating initial request below: When Categorical is used for simple problems (comparing one or two things) this is not much of an issue but we are seeing an increase in the complexity of studies where multiple products are being tested and often under multiple conditions/groupings. Multiplicity adjustments are increasingly needed here.Re: Multiple comparisons adjustments added to the Red Triangle for CategoricalHi, checking to see whether anything (even a simple pass-along to another platform) will be done in this space for JMP 14.
Thanks!
Re: Multiple comparisons adjustments added to the Red Triangle for CategoricalThank you. One option for implementation that I discussed with Chris G. when he visited us would be if a new option was added to the Red Triangle menu in Categorical which would launch a mainstream platform (GenReg?) with the Y variables from the side and the appropriate X variable or interaction of X's as read from the top of the Categorical grid, turn off most if not all of the model output there, and automatically trigger the Multiple Comparisons option from that mainstream platform. When that happens, the usual dialog would come up asking us to confirm the X effect and identify which type of comparison we want (Student's, Tukey, Dunnett, or ANOM). I'm sure there are other options but this sounded relatively simple to me and it would meet our requirement.
Re: Multiple comparisons adjustments added to the Red Triangle for CategoricalThere is ticket open in the JMP 14 Beta Feedback: No. SAS 7612211463
Re: Multiple comparisons adjustments added to the Red Triangle for CategoricalWhen Categorical is used for simple problems (comparing one or two things) this is not much of an issue but we are seeing an increase in the complexity of studies where multiple products are being tested and often under multiple conditions/groupings. Multiplicity adjustments are increasingly needed here.
