Show Student Project Implementation of the GUIDE algorithm for R Foundation for Statistical Computing
| Student Name: | Eva Hoermann |
| Mentor Name: | Torsten Hothorn |
| Title: | Implementation of the GUIDE algorithm |
| Abstract: |
In my project I plan to implement the statistical classification and regression tree algorithm GUIDE [1] based on the existing partykit infrastructure on R-Forge. GUIDE has negligible bias in split variable selection and is sensitive to curvature and local pairwise interactions between regressor variables, only to mention a few advantages. Since GUIDE is not implemented in R yet, my project aims to make it available to the R community.
[1] http://www.stat.wisc.edu/~loh/guide.html |
| Public info: |
I implemented GUIDE for regression trees as proposed by Wei-Yin Loh in "Regression trees with unbiased variable selection and interaction detection", Statistica Sinica 12 (2002), 361-386.
http://r-forge.r-project.org/plugins/scmsvn/viewcvs.php?root=partykit&view=rev&rev=139
http://r-forge.r-project.org/plugins/scmsvn/viewcvs.php?root=partykit&view=rev&rev=140 |
| Additional info: | http://r-forge.r-project.org/projects/partykit/ |
