Neolithic Programming Machine

    compute-quantile
    DefinitelyTyped icon, indicating that this package has TypeScript declarations provided by the separate @types/compute-quantile package

    1.0.1 • Public • Published

    Quantile

    NPM version Build Status Coverage Status Dependencies

    Computes a quantile for a numeric array.

    This module determines the nearest rank and returns the array value corresponding to that rank.

    This module does not currently support multiple interpolation methods, and instead uses a standard elementary method for estimating the quantile.

    Installation

    $ npm install compute-quantile

    For use in the browser, use browserify.

    Usage

    To use the module,

    var quantile = require( 'compute-quantile' );

    quantile( arr, p[, opts] )

    Given a probability 0 <= p <= 1, computes the quantile for an input array.

    var unsorted = [ 4, 3, 5, 1, 2 ];
     
    var q = quantile( unsorted, 0.25 );
    // returns 2

    If the input array is already sorted in ascending order, you can set the sorted option to true.

    var sorted = [ 1, 2, 3, 4, 5 ];
     
    var q = quantile( sorted, 0.25, { 'sorted': true } );
    // returns 2

    Examples

    var quantile = require( 'compute-quantile' );
     
    // Simulate some data...
    var data = new Array( 1000 );
     
    for ( var i = 0, len = data.length; i < len; i++ ) {
        data[ i ] = Math.round( Math.random()*1000 );
    }
     
    // Compute the p-quantile for p = 0.5 (median):
    console.log( quantile( data, 0.5 ) );

    To run the example code from the top-level application directory,

    $ node ./examples/index.js

    Notes

    The probability p is equal to k / q, where q is the total number of quantiles and k is the k th quantile.

    If the input array is not sorted in ascending order, the function is O( N log(N) ), where N is the input array length. If the array is sorted, the function is O(1).

    Tests

    Unit

    Unit tests use the Mocha test framework with Chai assertions. To run the tests, execute the following command in the top-level application directory:

    $ make test

    All new feature development should have corresponding unit tests to validate correct functionality.

    Test Coverage

    This repository uses Istanbul as its code coverage tool. To generate a test coverage report, execute the following command in the top-level application directory:

    $ make test-cov

    Istanbul creates a ./reports/coverage directory. To access an HTML version of the report,

    $ make view-cov

    License

    MIT license.


    Copyright

    Copyright © 2014. Athan Reines.

    Install

    npm i compute-quantile

    DownloadsWeekly Downloads

    2,137

    Version

    1.0.1

    License

    none

    Last publish

    Collaborators

    • kgryte