Approximate Marginal Likelihood for Gaussian Models

AMLGM provides efficient computational tools for approximate marginal likelihood estimation and Bayesian inference in Gaussian and latent Gaussian models. Built to complement the R-INLA ecosystem with scalable, reproducible, and research-oriented workflows.

Get Started View Documentation

Quick Start

install.packages("INLA")

library(INLA)
library(amlgm)

result <- amlgm(...)
summary(result)

Why AMLGM?

Designed for modern Bayesian statistical modelling.

Fast

Efficient numerical approximations for complex Gaussian models without computationally intensive sampling.

Reliable

Robust estimation procedures suitable for large datasets and high-dimensional models.

Integrated

Compatible with the R-INLA ecosystem and existing Bayesian workflows.

Key Features

Approximate Marginal Likelihood

Efficient computation for model comparison and Bayesian evidence.

Latent Gaussian Models

Designed to work naturally with Gaussian latent structures.

Model Diagnostics

Posterior summaries, uncertainty estimates, and predictive evaluation.

Open Source

Developed for reproducible statistical research.

Start Building Bayesian Models Today

Explore documentation, tutorials, and examples to begin using AMLGM in your research.


Documentation