This function predicts values based on generalized boosted model objects.
Usage
forecast_by_gbm(
object,
newx,
type = c("response", "class", "link"),
newoffset,
pthr = 0.5,
...
)Arguments
- object
an object of class
gbm.object- newx
the new values at which prediction are to be made
- type
the type of prediction. Three options are available
"response"predicted values are on the same scale as the outcome, e.g. for Bernoulli the returned values are probabilities, for Poisson the expected counts
"class"the class corresponding to the max probability
"link"predictions are on the scale of f(x), e.g. for Bernoulli the returned value is on the log odds scale
- newoffset
the new offset, if provided it is added to the predicted value
- pthr
probability threshold, used when
type = "class"and response is binomial to assign label to observation- ...
further arguments to
predict.gbm