Module hypothesis 

Hypothesis tests — real p-values from the distributions library, zero ad-hoc thresholds.

This replaces the prior one_sample_t, whose “p-value” was a placeholder (|t| > 1.96 ⇒ 0.05 else 0.1) and whose CI used the fixed 1.96 normal critical value regardless of sample size. Every test here draws its tail probability from the exact Student-t / F / χ² CDF, and t-CIs use the t critical value for the actual degrees of freedom.

Submodules (PROJECT RULE §11): t_tests (one-sample / paired / two-sample), anova (one-way F-test), chi_square (goodness-of-fit / independence). Specialized libraries map these results onto their own domain result types.

Re-exports

pub use anova::one_way_anova;

pub use anova::AnovaResult;

pub use chi_square::chi_square_gof;

pub use chi_square::chi_square_independence;

pub use chi_square::ChiSquareResult;

pub use nonparametric::friedman;

pub use nonparametric::ks_1sample;

pub use nonparametric::mann_whitney_u;

pub use nonparametric::mcnemar;

pub use nonparametric::FriedmanResult;

pub use nonparametric::KolmogorovSmirnovResult;

pub use nonparametric::MannWhitneyResult;

pub use nonparametric::NonparametricResult;

pub use t_tests::one_sample_t;

pub use t_tests::paired_t;

pub use t_tests::two_sample_t;

pub use t_tests::TTest;

pub use t_tests::TwoSampleTTest;

Modules

anova

One-way ANOVA — the F-test for equality of k group means, with a real F-tail p-value from fisher_f.

chi_square

χ² tests — goodness-of-fit and independence (contingency table), with a real upper-tail p-value from chi_squared.

nonparametric

Nonparametric & multiple-comparison tests (CI-SKM ch 6) — distribution-free tests for paired and multi-group comparisons, the standard tools for comparing classifiers/estimators across datasets. Real p-values from the χ²/F CDFs in distributions; within-block ranking reuses the statistics ranker.

t_tests

t-tests — one-sample, paired, and two-sample (pooled + Welch). Real Student-t p-values and t-based confidence intervals.