threshr: Threshold Selection and Uncertainty for Extreme Value Analysis version 1.0.6

threshr 1.0.6

Bug fixes and minor improvements

• Fixed 3 \link{} targets with missing (revdbayes) package anchors in the Rd file for ithresh().

threshr: Threshold Selection and Uncertainty for Extreme Value Analysis version 1.0.5

threshr 1.0.5

Bug fixes

• Fixed issues with the incorrect use of \itemize in some Rd files.

threshr: Threshold Selection and Uncertainty for Extreme Value Analysis version 1.0.4

threshr 1.0.4

Bug fixes and minor improvements

• Create the help file for the package correctly, with alias threshr-package.

• Activated 3rd edition of the testthat package

• README.md: Used app.codecov.io as base for codecov link.

threshr: Threshold Selection and Uncertainty for Extreme Value Analysis version 1.0.3

threshr 1.0.3

Bug fixes and minor improvements

• Tests in test-box_cox.R and test-inv_box_cox.R have been modified to avoid errors in the upcoming new release of the testthat package.

threshr: Threshold Selection and Uncertainty for Extreme Value Analysis version 1.0.2

threshr 1.0.2

Bug fixes and minor improvements

• In ithresh() a user-supplied (log-)prior R function can now be set for the binomial probability p of threshold exceedance. The functionality requires at least version 1.3.4 of the revdbayes package.

• A print method for class ithresh has been added.

• In plot.ithresh() a more informative error message is given if an inappropriate value of the argument which_v is supplied.

• In predict.ithresh() further arguments can now be passed to revdbayes::predict.evpost. In particular, the level(s) of predictive intervals can be set. An example has been added to the documentation.

• pkgdown documentation at https://paulnorthrop.github.io/threshr/

• revdbayes:: is used instead of revdbayes::: to avoid CRAN package check NOTEs.

threshr: Threshold Selection and Uncertainty for Extreme Value Analysis

threshr 1.0.1

Bug fixes and minor improvements

• Some examples and tests are modified slightly to avoid using unrealistically high or low thresholds.

• Dependence on R version changed to 3.3.0 to avoid CRAN NOTE.

Threshold Selection and Uncertainty for Extreme Value Analysis version 1.0.0

threshr: Threshold Selection and Uncertainty for Extreme Value Analysis

What does threshr do?

The threshr package deals primarily with the selection of thresholds for use in extreme value models. It also performs predictive inferences about future extreme values. These inferences can either be based on a single threshold or on a weighted average of inferences from multiple thresholds. The weighting reflects an estimated measure of the predictive performance of the threshold and can incorporate prior probabilities supplied by a user. At the moment only the simplest case, where the data can be treated as independent identically distributed observations, is considered, as described in Northrop et al. (2017). Future releases will tackle more general situations.

A simple example

The main function in the threshr package is ithresh. It uses Bayesian leave-one-out cross-validation to compare the extreme value predictive ability resulting from the use of each of a user-supplied set of thresholds. The following code produces a threshold diagnostic plot using a dataset gom containing 315 storm peak significant waveheights. We set a vector u_vec of thresholds; call ithresh, supplying the data and thresholds; and use then plot the results. In this minimal example (ithresh has further arguments) thresholds are judged in terms of the quality of prediction of whether the validation observation lies above the highest threshold in u_vec and, if it does, how much it exceeds this highest threshold.

u_vec_gom <- quantile(gom, probs = seq(0, 0.95, by = 0.05))
gom_cv <- ithresh(data = gom, u_vec = u_vec_gom)
plot(gom_cv)

Installation

To get the current released version from CRAN:

install.packages("threshr")

Vignette

See vignette("threshr-vignette", package = "threshr") for an overview of the package.