KOBT: Knockoff Boosted Tree
A novel strategy for conducting variable selection without prior model topology
knowledge using the knockoff method (Barber and Candes (2015) <doi:10.1214/15-AOS1337>)
with extreme boosted tree models (Chen and Guestrin (2016) <doi:10.1145/2939672.2939785>).
This method is inspired by the original knockoff method, where the differences between
original and knockoff variables are used for variable selection with false discovery rate
control. In addition to the original knockoff generating methods, two new sampling methods
are available to be implemented, namely the sparse covariance and principal component
knockoff methods. As results, the indices of selected variables are returned.
Please use the canonical form
to link to this page.