A comprehensive Julia package implementing a wide range of statistical hypothesis tests for data analysis.
HypothesisTests.jl is a Julia package that implements a wide range of statistical hypothesis tests for data analysis. It provides tools for performing parametric tests (like t-tests) and non-parametric tests (like Mann-Whitney U test) to evaluate statistical significance. The package calculates p-values, confidence intervals, and test statistics to support rigorous statistical inference.
Data scientists, researchers, and statisticians working in Julia who need to perform hypothesis testing as part of their data analysis workflows.
It offers a comprehensive, native Julia implementation of statistical tests with a clean API, eliminating the need to interface with external statistical software. The package integrates seamlessly with the JuliaStats ecosystem and provides both parametric and non-parametric testing options.
Hypothesis tests for Julia
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Implements both parametric (e.g., OneSampleTTest) and non-parametric tests (e.g., MannWhitneyUTest), covering a wide range of statistical scenarios as shown in the quick start examples.
Provides p-values, confidence intervals, and test statistics like t-statistic and degrees of freedom, enabling detailed inference and reporting directly from the API.
Supports one-tailed and two-tailed tests for directional analysis, demonstrated with the tail=:left or :right options in the pvalue and confint functions.
Designed to work naturally with Julia's data structures and statistical workflows, as emphasized in the package's philosophy and quick start snippets.
Primarily focuses on classical frequentist hypothesis tests and does not include Bayesian alternatives, which might be a limitation for modern statistical workflows.
Being a Julia-specific package, it requires adoption of the Julia ecosystem, which may not be feasible for teams standardized on other statistical platforms like Python or R.
While documentation badges are present, it might lack extensive tutorials or real-world case studies compared to more established statistical software packages.