Package: ngme2 1.0.1

ngme2: Linear Latent Non-Gaussian Models with Flexible Distributions

Fits and analyzes linear latent non-Gaussian models for temporal, spatial, and space-time data. The package provides model components for autoregressive and Ornstein-Uhlenbeck processes, random walks, Matern fields based on stochastic partial differential equations, separable and non-separable space-time models, graph-based Matern models, bivariate type-G fields, and user-defined sparse operators. Latent fields and observation models can use Gaussian and non-Gaussian noise distributions, including normal inverse Gaussian, generalized asymmetric Laplace, and skew-t distributions. Functions are included for simulation, likelihood-based estimation, prediction, cross-validation, convergence diagnostics, stochastic gradient optimization, batch-means confidence intervals, and posterior-like sampling. The modeling framework is described in Bolin, Jin, Simas and Wallin (2026) "A Unified and Computationally Efficient Non-Gaussian Statistical Modeling Framework" <doi:10.48550/arXiv.2602.23987>.

Authors:David Bolin [aut, cph], Xiaotian Jin [aut, cre], Alexandre Simas [aut], Jonas Wallin [aut], Andrea V. Rocha [ctb], Timothy A. Davis [ctb, cph], Patrick R. Amestoy [ctb, cph], Iain S. Duff [ctb, cph], John K. Reid [ctb, cph], Yanqing Chen [ctb, cph], Sivasankaran Rajamanickam [ctb, cph], Stefan Larimore [ctb, cph], William W. Hager [ctb, cph], University of Florida [cph], Regents of the University of Minnesota [cph], Free Software Foundation, Inc. [cph], Makoto Matsumoto [ctb, cph], Takuji Nishimura [ctb, cph], Alexander Chemeris [ctb, cph]