awildberg / distributions

loosely port of rubystats to crystal with bugfixes

distributions

loosely port of rubystats https://github.com/phillbaker/rubystats to crystal with bugfixes

Installation

manual

Find your crystal installation and the corresponding src directory:

crystal_bin="$(ls $(which crystal))"      # e.g. /Users/user/crystal/bin/crystal
crystal_dir="${crystal_bin%/*/*}/src/"    # gives /Users/user/crystal/src/

Place the main file distributions.cr and the libraries distributions into the src directory:

cp -R distributions.cr distributions "${crystal_dir}"

automatic

Simply put the following dependency in your shard.yml

dependencies:
  distributions:
    github: awildberg/distributions

shard install

Usage

require "distributions"

available classes

NormalDistribution
BinomialDistribution
BetaDistribution
CauchyDistribution
ExponentialDistribution
FishersExactTest
GammaDistribution
LognormalDistribution
PoissonDistribution
StudentTDistribution
UniformDistribution
WeibullDistribution

added after Matrix and Vector have been implemented:
MultivariateNormalDistribution

create object

norm  = Normal.new  2.0,  5       # shortcut for "NormalDistribution.new"
bino  = Binomial.new 2, 0.4
beta  = Beta.new  0.1,  0.3
cach  = Cauchy.new    2,  3
expo  = Exponential.new
fi    = Fisher.new
gamma = Gamma.new
log   = Lognormal.new
pois  = Poisson.new 10
st    = Student.new
uni   = Uniform.new 0, 10
wbull = Weibull.new

probability density function

puts norm.pdf 2.4                # 0.07953354110232176
puts norm.pdf [3.0, 3.3]         # [0.07820853879509115, 0.0771366738383632]
puts bino.pdf [0.2, 0.5]         # [0.4313110348439398, 0.49900595278955273]
puts beta.pdf 0.3                # 0.29564825937025924
puts cach.pdf 1.2                # 0.09905909321072323
puts expo.pdf 0.3                # 0.7408182206817179
puts gamma.pdf 0.8               # 0.44932896411722156
puts log.pdf 0.5                 # 0.6274960771159243
puts pois.pdf 2                  # 0.0022699964881242427
puts st.pdf 9                    # 0.0038818278802901303
puts uni.pdf 2.3                 # 0.1
puts wbull.pdf 2.3               # 0.10025884372280375

cumulative distribution function

puts norm.cdf 2.0                # 0.5
puts norm.cdf [2.3, 3.4]         # [0.5239221826541068, 0.6102612475557971]
puts bino.cdf [0.5, 0.9]         # [0.36, 0.36]
puts beta.cdf 0.9                # 0.8669336800281993
puts cach.cdf 3.2                # 0.6211189415908435
puts expo.cdf 1                  # 0.6321205588285577
puts gamma.cdf 0.4               # 0.32967995396436056
puts log.cdf 0.9                 # 0.4580448727856589
puts pois.cdf 10                 # 0.5830397501929856
puts st.pdf 9                    # -0.45482278617131633
puts uni.cdf 4                   # 0.4
puts wbull.cdf 3.2               # 0.9592377960216338

inverse cumulative distribution function

puts norm.icdf 0.6               # 3.2667355156789992
puts norm.icdf [0.1, 0.5]        # [-4.407757827723003, 2.0]
puts bino.icdf [0.9, 0.99]       # [2, 2]
puts beta.icdf 0.1               # 1.2091010943344349e-9
puts cach.icdf 0.5               # [2.0]
puts expo.icdf 0.8               # 1.6094379124341005
puts pois.icdf 0.999999          # 28
puts uni.icdf 0.7                # 7.0
puts wbull.icdf 0.9              # 2.302585092994046

random number

puts norm.rng 5                  # [0.989068826407212, 0.3864366469182574, -11.171031854542266, -2.382879260518245, -0.44308602711354883]
puts bino.rng 5                  # [1.0, 0.0, 0.0, 1.0, 2.0]
puts beta.rng 5                  # [0.4843079351756574, 0.0003248099619832699, 0.9920658717948483, 2.7869013274498478e-5, 0.13667412951628483]
puts cach.rng                    # 12.123106697726463
puts expo.rng                    # 0.8030553422389787
puts gamma.rng                   # 0.31707382996802885
puts log.rng                     # 0.6366757373839796
puts pois.rng                    # 14
puts st.rng                      # -0.6552081722716109
puts uni.rng 5                   # [9.870949403391021, 7.955413730795023, 3.8598986539463893, 3.6485901595039554, 4.968370145159594]
puts wbull.rng                   # 0.2298211759745445

fishers exact test

puts fi.exact 2.3, 4.2, 1.3, 5.3 # 0.4389521037471969

distributions's Issues

Adding MultivariateNormalDistribution

Needs Matrix and Vector Objects

Add more array return to rng functions