Vari

A lightweight probability and statistics library for dealing with various distributions.

Creating a distribution

We bring Vari into the project.

import Vari from 'vari'

We can then create a distribution. Functions return a distribution object.

// Creates a Normal Distribution (return)
const distribution = Vari.Normal(mean, variance)

We can then access any information we want in the distribution.

distribution.mean // expected value
distribution.sd // standard deviation of distribution
distribution.pdf(x) // returns a probability density
distribution.cdf(x) // returns a cumulutive density

Distributions included (With API):

const Binomial = Vari.B(n, p)
const Hypergeometric = Vari.H(N, K, n)
const Poisson = Vari.P(lamda)
const Normal = Vari.N(mean, variance)
const Exponential = Vari.Exp(lamda)
const Uniform = Vari.U(a, b)

Other functions

const expectedValue = Vari.E(arr) // takes an array
const binomialCoefficient = Vari.ncr(5,3) // returns 10
const variance = Vari.Var(arr) // takes an array

Vari accomodates a lot of notation styles:

// Expected Value
Vari.E(arr) || Vari.mean(arr)
// Distributions (either full name or mathematical symbol)
Vari.B(n, p) || Vari.Binomial(n, p) //returns binomial distribution