FSharp.Stats is a multipurpose project for statistical testing, linear algebra, machine learning, fitting and signal processing.

Amongst others, following functionalities are covered:

Descriptive statistics

• Measures of central tendency
• Measures of dispersion
• Correlation
• Quantile/Rank
• Distribution

Fitting

• Linear regression
  • Simple linear regression (weighted and constrained)
  • Polynomial regression (weighted and constrained)
• Nonlinear regression
• Spline regression
• Goodness of fit

Interpolation

• Linear spline interpolation
• Polynomial interpolation
• Cubic spline interpolation
• Akima subspline interpolation
• Hermite subspline interpolation

Signal processing

• Continuous wavelet transform
• Smoothing filters
• Peak detection

Linear Algebra

• Singular value decomposition

Machine learning

• PCA
• Clustering
• Surprisal analysis

Optimization

• Brent minimization
• Bisection
• Nelder Mead

Statistical testing

• t test, H test, etc.
  
• ANOVA
  
• Post hoc tests
  
• q values
  
• SAM
  
• RMT

Documentation

Indepth explanations, tutorials and general information about the project can be found here or at fslab. The documentation and tutorials for this library are automatically generated (using the F# Formatting) from *.fsx and *.md files in the docs folder. If you find a typo, please submit a pull request!

Contributing

Please refer to the Contribution guidelines.

Development

to build the project, run either build.cmd or build.sh depending on your OS.

build targets are defined in the modules of /build/build.fsproj.

Some interesting targets may be:

• ./build.cmd runtests will build the project and run tests
• ./build.cmd watchdocs will build the project, run tests, and build and host a local version of the documentation.
• ./build.cmd release will start the full release pipeline.

Library license

The library is available under Apache 2.0. For more information see the License file in the GitHub repository.

Citation

FSharp.Stats can be cited using its zenodo record.

Benedikt Venn, Lukas Weil, Kevin Schneider, David Zimmer & Timo Mühlhaus. (2022). fslaborg/FSharp.Stats. Zenodo. https://doi.org/10.5281/zenodo.6337056