margEVT: Regularized Point Processes and Stochastic Marginalization for Extremes

Implements a non-stationary extreme value analysis framework by coupling a covariate-driven non-homogeneous Poisson process (NHPP) with (adaptive) elastic-net regularization, analytical gradients and a lower bound on the shape parameter. Provides methods for estimating conditional return levels and unconditional (marginalized) return levels via parametric stochastic integration over stable vector autoregressive VAR(p) covariate trajectories, or non-parametric annual-block resampling. Supports block-specific penalty controls, BIC selection along a regularization path, Karush-Kuhn-Tucker convergence checks, time-rescaling residual diagnostics, conditional parametric bootstrap inference, and walk-forward assessment.

Version:	0.5.0
Imports:	graphics, grDevices, parallel, stats, utils
Suggests:	knitr, rmarkdown, testthat (≥ 3.0.0), vars
Published:	2026-10-23
DOI:	10.32614/CRAN.package.margEVT
Author:	Rodrigo Fonseca Villa [aut, cre]
Maintainer:	Rodrigo Fonseca Villa <rodrigo03.villa at gmail.com>
BugReports:	https://github.com/rodrigosqrt3/margEVT/issues
License:	GPL (≥ 3)
URL:	https://github.com/rodrigosqrt3/margEVT
NeedsCompilation:	no
Citation:	margEVT citation info
Materials:	README, NEWS
In views:	ExtremeValue
CRAN checks:	margEVT results

Documentation:

Reference manual:	margEVT.html , margEVT.pdf
Vignettes:	Covariate-driven extremes with margEVT (source, R code)

Downloads:

Package source:	margEVT_0.5.0.tar.gz
Windows binaries:	r-devel: margEVT_0.5.0.zip, r-release: margEVT_0.5.0.zip, r-oldrel: margEVT_0.5.0.zip
macOS binaries:	r-release (arm64): margEVT_0.5.0.tgz, r-oldrel (arm64): margEVT_0.5.0.tgz, r-release (x86_64): margEVT_0.5.0.tgz, r-oldrel (x86_64): margEVT_0.5.0.tgz
Old sources:	margEVT archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=margEVT to link to this page.