ens_crps: Continuous Rank Probability Score (CRPS) for an ensemble.

View source: R/ens_crps.R

ens_crpsR Documentation

Continuous Rank Probability Score (CRPS) for an ensemble.

Description

The CRPS and its decomposition are computed as columns in a harp_fcst object. Typically the scores are aggregated over lead time, but other grouping variables cam be chosen.

Usage

ens_crps(
  .fcst,
  parameter,
  groupings = "leadtime",
  num_ref_members = NA,
  keep_full_output = FALSE,
  show_progress = FALSE
)

Arguments

.fcst

A harp_fcst object with tables that have a column for observations, or a single forecast table.

parameter

The name of the column for the observed data.

groupings

The groups for which to compute the ensemble mean and spread. See group_by for more information of how grouping works.

num_ref_members

The number of members for which to compute the fair CRPS.

keep_full_ouput

Logical. Whether to keep the full output to computing CRPS for ungrouped data.

Value

An object of the same format as the inputs but with data grouped for the groupings column(s) and columns for crps, crps_pot and crps_rel.


andrew-MET/harpPoint documentation built on Feb. 23, 2023, 1:06 a.m.