# Vecchia-Laplace approximations of generalized Gaussian processes for big non-Gaussian spatial data

## TL;DR

This paper introduces a scalable and accurate approximation method for generalized Gaussian processes that efficiently handles large non-Gaussian spatial datasets by combining Laplace and Vecchia approximations.

## Contribution

It proposes a novel Vecchia-Laplace approximation for GGPs, enabling scalable inference for large non-Gaussian spatial data sets.

## Key findings

- The method is computationally efficient for large datasets.
- It provides accurate inference compared to existing methods.
- Implemented in an accessible R package.

## Abstract

Generalized Gaussian processes (GGPs) are highly flexible models that combine latent GPs with potentially non-Gaussian likelihoods from the exponential family. GGPs can be used in a variety of settings, including GP classification, nonparametric count regression, modeling non-Gaussian spatial data, and analyzing point patterns. However, inference for GGPs can be analytically intractable, and large datasets pose computational challenges due to the inversion of the GP covariance matrix. We propose a Vecchia-Laplace approximation for GGPs, which combines a Laplace approximation to the non-Gaussian likelihood with a computationally efficient Vecchia approximation to the GP, resulting in a simple, general, scalable, and accurate methodology. We provide numerical studies and comparisons on simulated and real spatial data. Our methods are implemented in a freely available R package.