numra-stats
Statistics for the Numra workspace — probability distributions, descriptive statistics, hypothesis tests, regression, and correlation.
Eleven distributions (continuous and discrete) with PDF / CDF / quantile / sample, a full descriptive suite, the standard tests (t, chi², KS, ANOVA), and three regression flavors.
Example
use ;
let samples = vec!;
let m = mean.unwrap;
let s = std_dev.unwrap;
// One-sample t-test against H0: μ = 5.0 at α = 0.05
let t = ttest_1samp.unwrap;
assert!; // Cannot reject H0
What's in this crate
Distributions (each with pdf, cdf, quantile, sample, mean, variance):
- Continuous:
Normal,Uniform,Exponential,GammaDist,BetaDist,ChiSquared,StudentT,FDist,LogNormal - Discrete:
Poisson,Binomial
Descriptive: mean, median, percentile, variance, std_dev, skewness, kurtosis, covariance, covariance_matrix
Hypothesis testing: ttest_1samp, ttest_ind, ttest_rel, chi2_test, ks_test, anova_oneway, TestResult
Regression: linregress, multiple_linregress, polyfit, RegressionResult
Correlation: pearson_r, spearman_r
Composes with
numra-ode— trajectory statistics and Monte Carlo summariesnumra-sde— SDE ensemble mean / variance / percentilenumra-pde— spatial-distribution statistics at the final timenumra-fit— regression as a statistical modelnumra-special— distribution PDFs back ontogamma,beta,erf
See interop workflows for verified PDE → statistics and stats-sampling → Monte Carlo ODE workflows.
Install
[]
= "0.1"
Or via the umbrella crate:
[]
= "0.1"
Documentation
- API: https://docs.rs/numra-stats
- Book: Descriptive stats · Distributions · Hypothesis testing · Regression
- Source: https://github.com/moussaoutlook/numra-rs/tree/main/numra-stats
License
Numra Academic & Research License (Non-Commercial). Academic and research use is free; commercial use requires a separate license — contact contact@spectralautomata.com. See LICENSE.