Generalized regression parameterization of nested effects
Sep 26, 2018 4:49 PM(170 views)
I'm using nested effects in the Generalized Regression and find that the parametrization seems strange, maybe wrong? In the parameter estimates I would expect one row for each level of the nesting factor, with just the name of that level in 's. Instead it uses the difference between that level and the reference level. For the last rows then you get an entries with [reference level - reference level]. That doesn't make sense to me. To illustrate run the attached code and you'll see rows in the parameter estimates table including these:
the subtraction of the reference category for State is fine by me, that's like how SAS parametrizes (I know that's different than other JMP platforms), but for the Region, it doesn't make sense? How do I interpret that?