Prediction from a TVGEVBayes Object — predict.TVGEVBayes • beverstan

Perform some computations about the predictive distribution from a TVGEVBayes object. For a "new" period of time given with newTimeRange, we can consider the maximum of the block maxima on this period. The distribution is not a GEV distribution in general, and due to the uncertainty on the parameters

Usage

# S3 method for class 'TVGEVBayes'
predict(
  object,
  newTimeRange = NULL,
  prob,
  type = "RL",
  approx = FALSE,
  trace = 0,
  ...
)

Arguments

object: A TVGEVBayes object.
newTimeRange: A character describing the "new" time range (period of time) as in timeRange.
prob: A vector of exceedance probabilities at which the predictive distribution is to be computed.
type: The type of result wanted. For now only the predictive Return Levels can be computed.
approx: Logical. If TRUE the predictive quantiles will be computed by using a linear interpolation method, while a series of uniroot is used instead.
trace: Integer level of verbosity.
...: Not used yet.

Examples

 
newTimeRange <- "2020_2026"