Regularized Point Processes and Stochastic Marginalization for Extremes R package margEVT version 0.3.0 CRAN Task View: ExtremeValue (https://CRAN.R-project.org/view=ExtremeValue) Implements a non-stationary extreme value analysis framework by coupling a covariate-driven Non-Homogeneous Poisson Process (NHPP) with Elastic-Net regularization and analytical gradients. Provides methods for estimating conditional return levels and unconditional (marginalized) return levels via parametric stochastic integration over stable Vector Autoregressive VAR(p) or univariate autoregressive covariate trajectories, or non-parametric annual-block resampling. Supports block-specific penalty controls, operational active-set thresholds, conditional parametric bootstrap inference, and walk-forward assessment. Software Imports: stats, utils, vars Suggests: testthat (>= 3.0.0) Rodrigo Fonseca Villa <rodrigo03.villa@gmail.com> Comprehensive R Archive Network (CRAN) Rodrigo Fonseca Villa [aut, cre] (ORCID: <https://orcid.org/0009-0005-2938-2270>) GPL (>= 3) 2026-09-23 application/tgz https://CRAN.R-project.org/package=margEVT doi:10.32614/CRAN.package.margEVT