Description Usage Arguments Details Value Examples
Given two datasets (observations and model), this function calculates a skill score for the agreement between the tails. The skill score is the area between the two PDFs above some threshold, weighted linearly outwards. This metric is based on Perkins, et al. 2013 in the Int. J. Climatology (doi: 10.1002/joc.3500).
1 |
obs |
A vector of observed data. |
mod |
A vector of model data. |
threshold |
The probability threshold where the tail begins. Default value: 0.95. |
... |
Arguments to the density estimation function. |
The upper tail is scored if the threshold value is >= 0.5;
otherwise, the lower tail is scored. The bkde
function
from the KernSmooth
package is used to estimate the PDFs of
the two datasets.
A numeric skill score; 1 indicates perfect agreement, 0 no agreement.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 | library(KernSmooth)
set.seed(22)
obs <- rgamma(1000, shape=1, scale=1)
perfect <- rgamma(1000, shape=1, scale=1)
good <- rgamma(1000, shape=6/5, scale=5/6)
bad <- rgamma(1000, shape=3, scale=1/3)
x <- namelist(obs, perfect, good, bad)
mplot(lapply(x, bkde), type="l", col=c("black","blue","green","red"), lty=1)
legend("topright", names(x), col=c("black","blue","green","red"), lty=1, lwd=2)
tailskill(obs, obs)
tailskill(obs, perfect)
tailskill(obs, good)
tailskill(obs, bad)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.