probability-distributions-k6
v0.1.1
Published
Generate random variables from a variety of probability distributions. Includes tools to shuffle an array or sample from it.
Downloads
11
Maintainers
Keywords
Readme
Probability Distributions Library for JavaScript
Functions for sampling random variables from probability distributions. Uses the same function names as R.
Adapted from https://github.com/Mattasher/probability-distributions in order to use its functions in https://k6.io/ scripts without any external dependencies.
Installation
npm install --save probability-distributions-k6
import PD from './node_modules/probability-distributions-k6/index.js';
Documentation and examples
See http://statisticsblog.com/probability-distributions/
Currently supported
Binomial distribution
Beta distribution
Cauchy distribution
Chi-Squared distribution
Exponential distribution
F distribution
Gamma distribution
Laplace distribution
Log Normal distribution
Negative Binomial distribution
Normal (Gaussian) distribution
Poisson distribution (not recommended for lambda > 100)
Sample (shuffle an array, or select items using optional array of weights)
Uniform distribution (with entropy option for standard uniform)
Uniform limited to whole numbers
Words (generate random words from a library of characters)
Visualization (show the values of a random variable in an animated loop)
Warning
This package contains additional distributions marked as "experimental". Use these with extreme caution.
License
MIT