semicontinuous_models 0.1.1

Reusable models for semi-continuous outcomes (two-part, Tweedie, and log-normal)
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57

#![forbid(unsafe_code)]

//! # `semicontinuous_models`
//!
//! Reusable models and diagnostics for non-negative semi-continuous outcomes:
//! two-part, Tweedie GLM, and log-normal with smearing.
//!
//! The crate was initially developed for healthcare outcome analyses, but the API is
//! intentionally domain-agnostic and can be reused in other settings.

pub mod input;
pub mod models;
pub mod preprocess;
pub mod utils;

pub use input::{InputError, ModelInput};
pub use preprocess::{
    OutcomeDiagnostics, column_has_variation, drop_constant_columns, outcome_diagnostics,
};
pub mod comparison {
    pub use crate::models::comparison::*;
}
pub mod matrix_ops {
    pub use crate::models::matrix_ops::*;
}

pub use models::two_part::{
    BootstrapOptions, BootstrapResult, BootstrapSummary, ConfidenceInterval, FitOptions,
    Regularization, TwoPartError, TwoPartModel, TwoPartPrediction, TwoPartReport,
    bootstrap as bootstrap_two_part, bootstrap_percentile_ci, bootstrap_summary,
    coefficient_confidence_intervals, fit_two_part_clustered_input, fit_two_part_input,
    fit_two_part_weighted_input, log_likelihood as two_part_log_likelihood,
};

pub use models::tweedie::{
    TweedieError, TweedieModel, TweedieOptions, TweediePrediction, TweedieReport,
    deviance as tweedie_deviance, fit_tweedie_input,
    quasi_log_likelihood as tweedie_quasi_log_likelihood,
};

pub use models::lognormal::{
    LogNormalError, LogNormalModel, LogNormalOptions, LogNormalPrediction, LogNormalReport,
    fit_lognormal_smearing_input, log_likelihood as lognormal_log_likelihood,
};

pub use models::selection::{
    CrossValidationError, CrossValidationOptions, CrossValidationResult, InformationCriteria,
    LogNormalCandidate, ModelFitMetrics, ParkTestResult, SelectionResult, TweedieCandidate,
    TweedieCvCandidate, compute_information_criteria, compute_model_fit_metrics,
    cross_validate_models_input, park_test, recommend_from_cv, select_models_input,
};

pub use models::comparison::{
    ComparisonTables, ModelComparison, ModelComparisonError, ModelComparisonOptions,
    ModelInformationCriteria, ModelScore, TweedieRankingRow, compare_models_input,
    render_comparison_tables,
};