Go to file
Arthur POULET 78025d09be
Dump version 0.2.3
2018-10-28 02:22:58 +01:00
spec Add median / quartiles 2018-10-20 15:26:35 +02:00
src Add median / quartiles 2018-10-20 15:26:35 +02:00
.gitignore Initialize the repository 2016-06-27 15:39:37 +02:00
.travis.yml Initialize the repository 2016-06-27 15:39:37 +02:00
CHANGELOG.md Improve normale distribution and update v0.1.6 2016-07-01 01:09:21 +02:00
LICENSE Change LICENSE to GPL-3.0 2017-04-13 22:42:05 +01:00
README.md Add median / quartiles 2018-10-20 15:26:35 +02:00
shard.yml Dump version 0.2.3 2018-10-28 02:22:58 +01:00

README.md

stats

An expressive implementation of statistical distributions. Compatible with crystal v0.26.1

Installation

Add this to your application's shard.yml:

dependencies:
  stats:
    github: Nephos/stats

Usage

Include it

require "stats"
include Stats

Normal distribution

NormaleDistribution.between # less_than, greater_than
  standard_deviation: 15,
  esperance: 100,
  min: 85,
  max: 115
  # => 0.6826894921370859

Binomial distribution

Math.binomial_distribution(
  tries: 3,
  probability: 0.5,
  success: 1)
  # => 0.375

Binomial coefficient

Math.coef_binomial(5, 2) # => 10

Factorial

Math.factorial(4) # => 24

Radian and degree

90.radian # => (Math::PI / 2)
(Math::PI / 4) # => 45.0

Series & Statistics

[1, 2, 3].mean # => 2.0
[1, 2, 3].variance # => 0.6667
[1, 2, 3].standard_deviation # => 0.8165
[1, 2, 3].quadratic_mean # => 2.16
[2, 32].geometric_mean # => 8.0
[40, 60].harmonic_mean # => 48.0
[1,2,3,2,1].macd 3 # => [2.0, 2.333, 2.0]

Correlations

[1,2,3,4].covariance [4,2,1,0] # => -1.625
[1,2,3,4].correlation_coef [1,2,3,3] + 1 > 1.5 # => true
[1,2,3,4].correlation_coef [-14,14,101,-100] + 1 > 1.5 # => false

Median

[1, 2, 5].median # => 2.0
[42, 1337].median # => 685.5

Quartiles & Boxplot

[1, 3, 5].first_quartile  # => 2.0 (alias of lower_quartile)
[1, 3, 5].second_quartile # => 3.0 (alias of median)
[1, 3, 5].third_quartile  # => 4.0 (alias of upper_quartile)
arr = [-23, -5, 2, 5, 5, 6, 7, 8, 14, 15, 42, 1337]

arr.first_quartile  # => 3.5  (Q1)
arr.second_quartile # => 6.5  (Q2)
arr.third_quartile  # => 14.5 (Q3)
arr.interquartile_range # => 11.0 (alias of iqr) (IQR = Q3 - Q1)

# Tukey's fences with k = 1.5 (default parameter value)
arr.lower_fence # => -13.0 (Q1 - 1.5 * IQR)
arr.upper_fence # => 31    (Q3 + 1.5 * IQR)
arr.lower_outliers # => [-23]
arr.upper_outliers # => [42, 1337]

# Tukey's fences with k = 3 for "far out" outliers
arr.upper_fence(3)    # => 47.5 (Q3 + 3 * IQR)
arr.upper_outliers(3) # => [1337]

Development

  • The lib is adapted to be usable with BigInt and BigFloat values

Contributing

  1. Fork it ( https://github.com/Nephos/stats/fork )
  2. Create your feature branch (git checkout -b my-new-feature)
  3. Commit your changes (git commit -am 'Add some feature')
  4. Push to the branch (git push origin my-new-feature)
  5. Create a new Pull Request

Contributors

  • Nephos Arthur Poulet - creator, maintainer