greeners-glm

Generalized linear models, discrete choice and related extensions.

Estimators / families

• GLM — generalized linear models with Family and Link.
• Logit / Probit — binary choice models.
• MNLogit — multinomial logit.
• OrderedLogit / OrderedProbit — ordinal models.
• Poisson / NegBin / NegBinP / GenPoisson — count models.
• ZINB / ZIP — zero-inflated count models.
• BetaModel — beta regression.
• GLMGam — generalized additive models.
• GEE — generalized estimating equations.
• ConditionalLogit / ConditionalMNLogit / ConditionalPoisson — conditional fixed-effects models.

Usage

[dependencies]
greeners-glm = "2.0"

use greeners_glm::{GLM, Family, Link};

let result = GLM::fit_with_link(
    &y, &x,
    Family::Gaussian,
    Link::Identity,
    Some(vec!["x1".to_string()]),
).unwrap();

Design notes

Common types such as Family, Link, GLM and Poisson are re-exported at the crate root. Module-specific result types stay under greeners_glm::<module>.