layer_gbm: Layer estimated using a gradient boosting model
In jonascrevecoeur/hirem: Hierarchical reserving models

Adds a new layer to the hierarchical reserving model. This layer will be estimated using the gbm package.

layer_gbm(
  obj,
  name,
  distribution,
  n.trees = 500,
  interaction.depth = 2,
  n.minobsinnode = 10,
  shrinkage = 0.1,
  bag.fraction = 0.5,
  cv.folds = 0,
  select_trees = "perf",
  filter = NULL,
  transformation = NULL
)

`obj`	The hierarchical reserving model
`name`	Character, name of the layer. This name should match the variable name in the data set
`distribution`	distribution argument passed to `gbm`
`n.trees`	n.trees argument passed to `gbm`
`interaction.depth`	interaction.depth argument passed to `gbm`
`n.minobsinnode`	n.minobsinnode argument passed to `gbm`
`shrinkage`	shrinkage argument passed to `gbm`
`bag.fraction`	bag.fraction argument passed to `gbm`
`select_trees`	Character string specifying the method for selecting the optimal number of trees after fitting the gbm "fixed": Use the number of trees specified in n.trees "perf": Update the number of trees using `gbm.perf`
`filter`	Function with input: Data set with same structure as the data passed to `hirem` output: TRUE/FALSE vector with same length as the number of rows in the input data set. FALSE indicates that this layer is zero for the current record.
`transformation`	Object of class `hirem_transformation` specifying the transformation applied before modelling this layer.