plot.error {randomSurvivalForest}R Documentation

Plot of Error Rate and Variable Importance

Description

Plot out-of-bag (OOB) error rate for the ensemble as a function of number of trees in the forest. Also plots importance values for predictors. Note this is the default plot method for the package.

Usage

    plot.error(x, ...)
    plot.rsf(x, ...)

Arguments

x An object of class (rsf, grow) or (rsf, predict).
... Further arguments passed to or from other methods.

Details

Plot of OOB error rate, with the b-th value being the error rate for the ensemble computed using the first b trees. Error rate is 1-C, where C is Harrell's concordance index. Rates given are between 0 and 1, with 0.5 representing the benchmark value of a procedure based on random guessing. A value of 0 is perfect.

In the orginal call if importance=TRUE (the default setting), then importance values for predictors will be plotted. A matrix with 3 columns is also printed. First column are importance values, second column are standardized importance values (divided by the maximum importance value), third column is the vector predictorWt. The importance value indicates how much misclassification increases, or decreases, for a new test case if the given predictor were not available for that case, adjusting for all other predictors used in growing the forest.

Author(s)

Hemant Ishwaran hemant.ishwaran@gmail.com and Udaya B. Kogalur ubk2101@columbia.edu

References

H. Ishwaran, Udaya B. Kogalur, Eugene H. Blackstone and Michael S. Lauer (2007). Random Survival Forests. Cleveland Clinic Technical Report.

L. Breiman (2001). Random forests, Machine Learning, 45:5-32.

F.E. Harrell et al. (1982). Evaluating the yield of medical tests, J. Amer. Med. Assoc., 247, 2543-2546.

See Also

rsf, predict.rsf.

Examples

  data(veteran, package = "randomSurvivalForest") 
  v.out <- rsf(Survrsf(time, status)~., veteran, ntree = 1000)
  plot.error(v.out, veteran)

[Package randomSurvivalForest version 3.0.1 Index]