gini-ss

0.2.1 • Public • Published

gini-ss

Greenkeeper badge Build Status

Compute the gini coefficient of the numeric data, with small sample correction.

 var giniSS = require('gini-ss');
 giniSS([0,0,0,0,1]) // --> 1.0
 giniSS([0,0,0]) // --> 0.0
 giniSS([3,3,3]) // --> 0.0
 giniSS([1,2,3,...,100]) // --> ~ 99/303 ~ 1/3

Small Sample Correction

The Gini coefficient with small sample correction has a value of 1.0 for the case of perfect inequality, when for example, with income data, all of the incomes are zero and one person has all the income.
The traditional Gini instead yields G = 1-(1/n) = (n-1)/n. The correction is simply multiplying by n/(n-1)

These converge as the number of samples n become large.

Background

For more information, see the Wikipedia article for Gini coefficient

Copyright

Copyright 2017 Paul Brewer, Economic and Financial Technology Consulting LLC

License

The MIT License

Package Sidebar

Install

npm i gini-ss

Weekly Downloads

8

Version

0.2.1

License

MIT

Unpacked Size

5.01 kB

Total Files

6

Last publish

Collaborators

  • drpaulbrewer