Conditional Variable Importance for Random Forests

Carolin Strobl, Anne-Laure Boulesteix, Thomas Kneib, Thomas Augustin, Achim Zeileis

Publikation: Wissenschaftliche FachzeitschriftOriginalbeitrag in FachzeitschriftBegutachtung

2 Downloads (Pure)

Abstract

Background Random forests are becoming increasingly popular in many scientific fields because they can cope with "small n large p" problems, complex interactions and even highly correlated predictor variables. Their variable importance measures have recently been suggested as screening tools for, e.g., gene expression studies. However, these variable importance measures show a bias towards correlated predictor variables. Results We identify two mechanisms responsible for this finding: (i) A preference for the selection of correlated predictors in the tree building process and (ii) an additional advantage for correlated predictor variables induced by the unconditional permutation scheme that is employed in the computation of the variable importance measure. Based on these considerations we develop a new, conditional permutation scheme for the computation of the variable importance measure. Conclusion The resulting conditional variable importance reflects the true impact of each predictor variable more reliably than the original marginal approach.
OriginalspracheEnglisch
Seiten (von - bis)1 - 11
FachzeitschriftBMC Bioinformatics
Jahrgang9
Ausgabenummer307
DOIs
PublikationsstatusVeröffentlicht - 2008

Dieses zitieren