Skip to main content

solow_discrete/
lib.rs

1//! # solow-discrete
2//!
3//! Discrete-choice and count regression models estimated by maximum likelihood:
4//! [`Logit`], [`Probit`], and [`Poisson`]. Each model is fit with a full Newton
5//! step using analytic log-likelihood, score, and Hessian, converging to the true
6//! optimum so that results agree with the canonical reference to machine precision.
7//!
8//! ```
9//! use ndarray::{array, Array2};
10//! use solow_discrete::Logit;
11//!
12//! let mut x = Array2::<f64>::ones((5, 2));
13//! x.column_mut(1)
14//!     .assign(&array![0.1, -0.4, 1.2, 0.7, -1.1]);
15//! let y = array![0.0, 0.0, 1.0, 1.0, 0.0];
16//! let res = Logit::new(y, x).unwrap().fit().unwrap();
17//! assert!(res.converged);
18//! assert_eq!(res.params.len(), 2);
19//! ```
20
21mod conditional;
22mod genpoisson;
23mod logistic_cv;
24mod model;
25mod multinomial;
26mod negbin;
27mod ordered;
28mod summary;
29mod truncated;
30mod zip;
31
32pub use conditional::{ConditionalLogit, ConditionalModel, ConditionalPoisson, ConditionalResults};
33pub use genpoisson::{GeneralizedPoisson, GeneralizedPoissonResults};
34pub use logistic_cv::LogisticRegressionCV;
35pub use model::{DiscreteResults, Logit, Poisson, Probit};
36pub use multinomial::{MNLogit, MNLogitResults};
37pub use negbin::{NegativeBinomial, NegativeBinomialResults};
38pub use ordered::{Distr, OrderedModel, OrderedResults};
39pub use truncated::{
40    HurdleCountModel, HurdleCountResults, TruncatedLFPoisson, TruncatedLFPoissonResults,
41};
42pub use zip::{ZeroInflatedPoisson, ZeroInflatedPoissonResults};