In pengyuwei94/evreg: Extreme value regression modelling

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "README-"
)

evreg

Extreme Value Regression

What does evreg do?

The evreg package provides functions that model non-stationary extreme values using univariate extreme value regression modelling, while also performing variable selection to objectively determine the best model for a given dataset. This is achieved by using generalized linear modelling for each parameter, and by fitting models with maximum likelihood estimates.

A simple example

One of the main functions in the evreg package is gevreg, which performs the model fitting for generalized extreme value regression modelling. Then, forward_gevreg can be used to perform forward selection to objectively determine the best model. The following code fits a GEV regression model using the gevreg function for fremantle dataset and then uses forward_gevreg perform forward selection.

library(evreg)
data <- fremantle[ , -which(names(fremantle) %in% c("Year"))]
fit0 <- gevreg(SeaLevel, data)
forward_gevreg(fit0)

Installation

To get the current released version from CRAN:

install.packages("evreg")

Vignette

pengyuwei94/evreg documentation built on Aug. 29, 2019, 1:06 p.m.