greeners 1.5.2

High-performance econometrics with R/Python formulas. Two-Way Clustering, Marginal Effects (AME/MEM), HC1-4, IV Predictions, Categorical C(var), Polynomial I(x^2), Interactions, Diagnostics. OLS, IV/2SLS, DiD, Logit/Probit, Panel (FE/RE), Time Series (VAR/VECM), Quantile!
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

use statrs::distribution::{
    ChiSquared, Continuous, ContinuousCDF, FisherSnedecor, Normal, StudentsT,
};

/// p-valor bilateral da distribuição t de Student.
pub fn t_pvalue_two(t: f64, df: f64) -> f64 {
    let Ok(dist) = StudentsT::new(0.0, 1.0, df) else {
        return f64::NAN;
    };
    2.0 * (1.0 - dist.cdf(t.abs()))
}

/// Quantil da distribuição t de Student (inversa da CDF).
/// Ex.: `t_quantile(0.975, df)` devolve o t crítico para IC 95%.
pub fn t_quantile(p: f64, df: f64) -> f64 {
    let Ok(dist) = StudentsT::new(0.0, 1.0, df) else {
        return f64::NAN;
    };
    dist.inverse_cdf(p)
}

/// p-valor da distribuição qui-quadrado.
pub fn chi2_pvalue(stat: f64, df: f64) -> f64 {
    let Ok(dist) = ChiSquared::new(df) else {
        return f64::NAN;
    };
    1.0 - dist.cdf(stat)
}

/// PDF da distribuição normal padrão N(0,1).
pub fn norm_pdf(x: f64) -> f64 {
    let Ok(dist) = Normal::new(0.0, 1.0) else {
        return f64::NAN;
    };
    dist.pdf(x)
}

/// Função logística (sigmoid): 1 / (1 + e^{-x}).
pub fn logistic(x: f64) -> f64 {
    1.0 / (1.0 + (-x).exp())
}

/// p-valor da distribuição F de Fisher-Snedecor (cauda superior).
pub fn f_pvalue(f: f64, df1: f64, df2: f64) -> f64 {
    let Ok(dist) = FisherSnedecor::new(df1, df2) else {
        return f64::NAN;
    };
    1.0 - dist.cdf(f)
}