Low commit activity in last 3 years
A long-lived project that still receives updates
High performance, native (C) implementations of various statistical functions.
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
 Dependencies

Development

Runtime

 Project Readme

Ruby Native Statistics

Test status

This is a native extension to Ruby that adds native (C) statistical functions to the Array class. At present the following functions are provided:

Check the Github Actions build to see the currently supported versions of Ruby. This list will match whatever stable versions are specified at https://www.ruby-lang.org/en/downloads/.

It is generally more performant than calculating these values with pure Ruby. For a comparison, run the benchmarks with rake benchmark.

Test (Ruby 3.3.0) Run 1 Run 2 Run 3 Run 4 Run 5
bench_native_stdev 0.000069 0.000074 0.000064 0.000065 0.000065
bench_ruby_stdev 0.000947 0.000932 0.000927 0.000948 0.000909
Test (Ruby 3.3.0) Run 1 Run 2 Run 3 Run 4 Run 5
bench_native_median 0.000719 0.00067 0.000659 0.000638 0.000668
bench_ruby_median 0.000774 0.000743 0.000724 0.000697 0.000683
Test (Ruby 3.3.0) Run 1 Run 2 Run 3 Run 4 Run 5
bench_native_mean 0.000035 0.000035 0.000034 0.000032 0.000033
bench_ruby_mean 0.000291 0.000287 0.000291 0.000299 0.000281

Found a bug? Need a function?

If you found a bug or need a particular function, please let me know! I work on this gem in my spare time, mainly for learning purposes. Feel free to open a PR or a Github issue and I'll take a look as soon as possible.

Usage

require 'ruby_native_statistics'
r = [1,3,21,32,42]

# calculate sample standard deviation, you can also use "stdevs"
p r.stdev

# calculate population standard deviation
p r.stdevp

# calculate mean
p r.mean

# calculate median
p r.median

# calculate percentile
p r.percentile(0.3333)

Implementation notes

Percentile

Percentile uses the same rounding method as Excel, sometimes called R7.

Links

This is the third version of this gem, and it is a total rewrite of a SWIG-based design. Lots of thanks to the following resources: