trimTrees
(Package: trimTrees) :
Trimmed Opinion Pools of Trees in Random Forest
This function creates point and probability forecasts from the trees in a random forest using Jose et al.'s trimmed opinion pool, a trimmed average of the trees' empirical cumulative distribution functions (cdf). For tuning purposes, the user can input the trimming level used in this trimmed average and then compare the scores of the trimmed and untrimmed opinion pools, or ensembles.
● Data Source:
CranContrib
● Keywords: classif, randomForest, regression, tree
● Alias: trimTrees
●
0 images
cinbag
(Package: trimTrees) :
Modified Classification and Regression with Random Forest
cinbag implements a modified random forest algorithm (based on the source code from the randomForest package by Andy Liaw and Matthew Wiener and on the original Fortran code by Leo Breiman and Adele Cutler) to return the number of times a row appears in a tree's bag. cinbag returns a randomForest object, e.g., rfobj, with an additional output, a matrix with inbag counts (rows) for each tree (columns). For instance, rfobj$inbagCount is similar to rfobj$inbag, but with inbag counts instead of inbag indicators.
● Data Source:
CranContrib
● Keywords: classif, randomForest, regression, tree
● Alias: cinbag
●
0 images
hitRate
(Package: trimTrees) :
Empirical Hit Rates for a Crowd of Forecasters
This function calculates the empirical hit rates for a crowd of forecasters over a testing set. The function takes as its arguments the forecasters' probability integral transform (PIT) values – one for each testing set row – and the prediction interval of interest.
● Data Source:
CranContrib
● Keywords: classif, randomForest, regression, tree
● Alias: hitRate
●
0 images
providing 
.