README.md

In YunyiShen/multiMaxEnt: Multispecies MaxEnt Model Via Gibbs Point Process

multiMaxEnt

An experimental R package to fit a multispecies extension of maxent (more specific maxnet implementation) using its equivalence with Poisson and Gibbs point process.

The idea behind

Fithian and Hastie 2014 revealed the equivalence between Phillips' MaxEnt model, a Logistic regression with infinitely many "background points" a Poisson point process whose natural multi-class extension is Gibbs point process. In the same year, Baddeley et al. proposed to use a similar equivalence between Logistic regression and a Gibbs point process which was implemented in the R package spatstat.core.

This experimental package aim to combine the two paper and test the idea of using Gibbs point process to fit a multispecies version of MaxEnt.

YunyiShen/multiMaxEnt documentation built on Dec. 18, 2021, 8:22 p.m.