Helper functions for computing the relative influence of each variable in the TDboost object.
relative.influence(object, n.trees)
permutation.test.TDboost(object, n.trees)
TDboost.loss(y,f,w,offset,dist,baseline)Returns an unprocessed vector of estimated relative influences.
a TDboost object created from an initial call to TDboost.
the number of trees to use for computations.
For TDboost.loss: These components are the
outcome, predicted value, observation weight, offset, distribution, and comparison
loss function, respectively.
Yi Yang yi.yang6@mcgill.ca, Wei Qian wxqsma@rit.edu and Hui Zou hzou@stat.umn.edu
This is not intended for end-user use. These functions offer the different
methods for computing the relative influence in summary.TDboost.
TDboost.loss is a helper function for permutation.test.TDboost.
Yang, Y., Qian, W. and Zou, H. (2013), “A Boosted Tweedie Compound Poisson Model for Insurance Premium” Preprint.
G. Ridgeway (1999). “The state of boosting,” Computing Science and Statistics 31:172-181.
J.H. Friedman (2001). "Greedy Function Approximation: A Gradient Boosting Machine," Annals of Statistics 29(5):1189-1232.
summary.TDboost