In the dataset that I am working on, we have a list of occupations and I want to use the MBR node to group like occupations together. However, the occupations variable is nominal and MBR only works with interval variables. Is there any way to use Princomp to make the occupations variable usable in the MBR node?
Why do you have to use the MBR node, especially since it doesn't seem to fit? Why couldn't some other method work here? What is it exactly that you wish this analysis to do? And I don't see how using Principal Components helps when the MBR method doesn't fit.