time-to-botec/squiggle/node_modules/@stdlib/stats/levene-test/README.md
NunoSempere b6addc7f05 feat: add the node modules
Necessary in order to clearly see the squiggle hotwiring.
2022-12-03 12:44:49 +00:00

4.5 KiB

Levene's Test

Compute Levene's test for equal variances.

Levene's test is used to test the null hypothesis that the variances of k groups are equal against the alternative that at least two of them are different.

Usage

var leveneTest = require( '@stdlib/stats/levene-test' );

leveneTest( x[, y, ..., z][, opts] )

Calculates Levene's test for input arrays x, y, ..., z holding numeric observations.

// Data from Hollander & Wolfe (1973), p. 116:
var x = [ 2.9, 3.0, 2.5, 2.6, 3.2 ];
var y = [ 3.8, 2.7, 4.0, 2.4 ];
var z = [ 2.8, 3.4, 3.7, 2.2, 2.0 ];

var out = leveneTest( x, y, z );
/* returns
    {
        'rejected': false,
        'alpha': 0.05,
        'df': [ 2, 11 ],
        'pValue': ~0.1733,
        'statistic': ~2.0638,
        ...
    }
*/

The function accepts the following options:

  • alpha: number on the interval [0,1] giving the significance level of the hypothesis test. Default: 0.05.
  • groups: an array of group indicators. Only applicable when providing a single numeric array holding all observations.

By default, the test is carried out at a significance level of 0.05. To test at a different significance level, set the alpha option.

var x = [ 2.9, 3.0, 2.5, 2.6, 3.2 ];
var y = [ 3.8, 2.7, 4.0, 2.4 ];
var z = [ 2.8, 3.4, 3.7, 2.2, 2.0 ];

var out = leveneTest( x, y, z, {
    'alpha': 0.01
});
/* returns
    {
        'rejected': false,
        'alpha': 0.01,
        'df': [ 2, 11 ],
        'pValue': ~0.1733,
        'statistic': ~2.0638,
        ...
    }
*/

In addition to providing multiple arrays, the function supports providing a single numeric array holding all observations along with an array of group indicators.

var arr = [
    2.9, 3.0, 2.5, 2.6, 3.2,
    3.8, 2.7, 4.0, 2.4,
    2.8, 3.4, 3.7, 2.2, 2.0
];
var groups = [
    'a', 'a', 'a', 'a', 'a',
    'b', 'b', 'b', 'b',
    'c', 'c', 'c', 'c', 'c'
];
var out = leveneTest( arr, {
    'groups': groups
});

The returned object comes with a .print() method which, when invoked, prints a formatted output of test results. The method accepts the following options:

  • digits: number of decimal digits displayed for the outputs. Default: 4.
  • decision: boolean indicating whether to print the test decision. Default: true.
var x = [ 2.9, 3.0, 2.5, 2.6, 3.2 ];
var y = [ 3.8, 2.7, 4.0, 2.4 ];
var z = [ 2.8, 3.4, 3.7, 2.2, 2.0 ];

var out = leveneTest( x, y, z );
console.log( out.print() );
/* =>
    Levene's test for Homogeneity of Variance

    Null hypothesis: The variances in all groups are the same.

        df 1: 2
        df 2: 11
        F score: 2.0638
        P Value: 0.1733

    Test Decision: Fail to reject null in favor of alternative at 5% significance level
*/

Examples

var leveneTest = require( '@stdlib/stats/levene-test' );

// Data from Hollander & Wolfe (1973), p. 116:
var x = [ 2.9, 3.0, 2.5, 2.6, 3.2 ];
var y = [ 3.8, 2.7, 4.0, 2.4 ];
var z = [ 2.8, 3.4, 3.7, 2.2, 2.0 ];

var out = leveneTest( x, y, z );
/* returns
    {
        'rejected': false,
        'alpha': 0.05,
        'df': [ 2, 11 ],
        'pValue': ~0.1733,
        'statistic': ~2.0638,
        ...
    }
*/

var table = out.print();
/* returns
    Levene's test for Homogeneity of Variance

    Null hypothesis: The variances in all groups are the same.

        df 1: 2
        df 2: 11
        F score: 2.0638
        P Value: 0.1733

    Test Decision: Fail to reject null in favor of alternative at 5% significance level
*/