Quotation Strobl, Carolin and Boulesteix, Anne-Laure and Kneib, Thomas and Augustin, Thomas and Zeileis, Achim. 2008. Conditional Variable Importance for Random Forests. BMC Bioinformatics 9 (307): S. 1-11.




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.


Press 'enter' for creating the tag

Publication's profile

Status of publication Published
Affiliation WU
Type of publication Journal article
Journal BMC Bioinformatics
Citation Index SCI
Language English
Title Conditional Variable Importance for Random Forests
Volume 9
Number 307
Year 2008
Page from 1
Page to 11
Reviewed? Y
URL http://www.biomedcentral.com/1471-2105/9/307
DOI http://dx.doi.org/10.1186/1471-2105-9-307


Zeileis, Achim (Former researcher)
Augustin, Thomas
Boulesteix, Anne-Laure
Kneib, Thomas
Strobl, Carolin
Institute for Statistics and Mathematics IN (Details)
Google Scholar: Search