Using neural networks to estimate parameters in spatial point process models

TL;DR

This paper introduces a neural network-based method for estimating all unknown parameters in spatial point process models from observed data, demonstrating improved accuracy over traditional methods in simulations and real data applications.

Contribution

The paper presents a novel neural network approach capable of estimating all parameters in spatial point process models, applicable to any model that can be simulated from.

Findings

The method accurately recovers parameters in simulation studies.

It outperforms traditional estimation methods in certain scenarios.

Successfully applied to real-world spatial data.

Abstract

In this paper, I show how neural networks can be used to simultaneously estimate all unknown parameters in a spatial point process model from an observed point pattern. The method can be applied to any point process model which it is possible to simulate from. Through a simulation study, I conclude that the method recovers parameters well and in some situations provide better estimates than the most commonly used methods. I also illustrate how the method can be used on a real data example.