R help - Mar 2006 - hierarchical partitioning

Hello,

I would like to know what it means, when the hier.part's output has a data
frame of a negative independent contribution for variables (?IJ) and how
this can be caused.

Thank you very much.

Yours sincerely

M. Heuner

Hello,

If you obtain negative values in the results of your hierarchical
partioning it means that the join contribution of your variables
explain more of the variation than the sum of the individual effects
of your variables.
You can find some more information in the original paper:
Chevan & Sutherland 1991 - Hierarchical partioning - The American
Statistician 45(2):90-96

Fran?ois Michonneau

Dear Meike,

a variable with negative entry for J in IJ (I cannot become negative) 
explains more in the joint model than it can explain when alone in the model. 
You can find discussions of such phenomena e.g. under the key 
word "suppression" or "suppressor variables".

And by the way, as you are using hier.part, you might also be interested in 
the package relaimpo (linear model with R-squared only, but more different 
metrics for partitioning).

Regards, Ulrike

****************************** 
Prof. Dr. Ulrike Gr?mping 
Fachbereich II 
TFH Berlin 
Luxemburger Str. 10 
13353 Berlin
www: www.tfh-berlin.de/~groemp/