cpi.plot: Central Prediction Intervals Plot

In jobstdavid/eppverification: Verification Tools for the Statistical Postprocessing of Ensemble Forecasts

Central Prediction Intervals Plot

Description

This function plots the central prediction intervals given evaluation points, observations of an univariate variable and quantiles of a predictive distribution for a certain interval range.

Usage

cpi.plot(
  x,
  y,
  lower,
  upper,
  nominal.coverage,
  x.lab = "",
  y.lab = "",
  title = paste0(nominal.coverage, "% Central Prediction Interval"),
  info = FALSE
)

Arguments

x	evaluation points of class "Date", "numeric" or "integer"
y	vector of observations
lower	vector with the lower quantiles (depending on nominal.coverage; see details)
upper	vector with the upper quantiles (depending on nominal.coverage; see details)
nominal.coverage	numeric; nominal coverage in % (see details)
x.lab	character; label of x-axis; default: ""
y.lab	character; label of y-axis; default: ""
title	character; title of the plot; default: "nominal.coverage% Central Prediction Interval"
info	logical; if TRUE the mean central prediction interval width and coverage are calculated for the plot (see details); if FALSE the mean central prediction interval width and coverage are not calculated; default: FALSE

Details

For x of length n, y, lower and upper should be given as vector of length n, where the i-th entry of x belongs to the i-th entry of the other vectors. NA's are omitted.

A lower central prediction interval width indicates a sharper forecast.

A score of the central prediction interval coverage close to nominal.coverage indicates a more calibrated forecast. A central prediction interval coverage score of nominal.coverage is optimal and indicates a calibrated forecast.

Value

ggplot object with a plot of the central prediction intervals.

Author(s)

David Jobst

References

Gneiting, T. and Raftery, A. (2007). Strictly Proper Scoring Rules, Prediction,and Estimation. Journal of the American Statistical Association, 102, 359-378.

Examples

# simulated data
n <- 30
x1 <- 1:n
x2 <- seq(Sys.Date(), by = "day", length.out = n)
y <- rnorm(n, mean = 1:n)
nominal.coverage <- 90
alpha <- (100-nominal.coverage)/100
lower <- qnorm(alpha/2, rnorm(n, mean = 1:n))
upper <- qnorm((1-alpha/2), rnorm(n, mean = 1:n))
x.lab <- "Date"
y.lab <- "Value"
title <- "Central Prediction Interval"


# cpi.plot plot
cpi.plot(x = x1, y = y, lower = lower, upper = upper,
nominal.coverage = nominal.coverage)
cpi.plot(x = x2, y = y, lower = lower, upper = upper,
nominal.coverage = nominal.coverage, x.lab = x.lab, y.lab = y.lab,
title = title, info = TRUE)

jobstdavid/eppverification documentation built on May 13, 2024, 5:20 p.m.