Correlation Test

Compute a Pearson product-moment correlation test between paired samples.

Installation

npm install @stdlib/stats-pcorrtest

Usage

var pcorrtest = require( '@stdlib/stats-pcorrtest' );

pcorrtest( x, y[, opts] )

By default, the function performs a t-test for the null hypothesis that the paired data in arrays or typed arrays x and y have a Pearson correlation coefficient of zero.

var x = [ 0.7, -1.6, -0.2, -1.2, -0.1, 3.4, 3.7, 0.8, 0.0, 2.0 ];
var y = [ 1.9, 0.8, 1.1, 0.1, -0.1, 4.4, 5.5, 1.6, 4.6, 3.4 ];

var out = pcorrtest( x, y );
/* e.g., returns
    {
        'alpha': 0.05,
        'rejected': true,
        'pValue': ~0.006,
        'statistic': ~3.709,
        'ci': [ ~0.332, ~0.95 ],
        'nullValue': 0,
        'pcorr': ~0.795,
        // ...
    }
*/

The returned object comes with a .print() method which when invoked will print a formatted output of the results of the hypothesis test. print accepts a digits option that controls the number of decimal digits displayed for the outputs and a decision option, which when set to false will hide the test decision.

console.log( out.print() );
/* e.g., =>
    t-test for Pearson correlation coefficient

    Alternative hypothesis: True correlation coefficient is not equal to 0

        pValue: 0.006
        statistic: 3.709
        95% confidence interval: [0.3315,0.9494]

    Test Decision: Reject null in favor of alternative at 5% significance level
*/

The function accepts the following options:

• alpha: number in the interval [0,1] giving the significance level of the hypothesis test. Default: 0.05.
• alternative: Either two-sided, less or greater. Indicates whether the alternative hypothesis is that x has a larger mean than y (greater), x has a smaller mean than y (less) or the means are the same (two-sided). Default: two-sided.
• rho: number denoting the correlation between the x and y variables under the null hypothesis. Default: 0.

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

var x = [ 0.7, -1.6, -0.2, -1.2, -0.1, 3.4, 3.7, 0.8, 0.0, 2.0 ];
var y = [ 1.9, 0.8, 1.1, 0.1, -0.1, 4.4, 5.5, 1.6, 4.6, 3.4 ];

var out = pcorrtest( x, y, {
    'alpha': 0.1
});
var table = out.print();
/* e.g., returns
    t-test for Pearson correlation coefficient

    Alternative hypothesis: True correlation coefficient is not equal to 0

        pValue: 0.006
        statistic: 3.709
        90% confidence interval: [0.433,0.9363]

    Test Decision: Reject null in favor of alternative at 10% significance level
*/

By default, a two-sided test is performed. To perform either of the one-sided tests, set the alternative option to less or greater.

var x = [ 0.7, -1.6, -0.2, -1.2, -0.1, 3.4, 3.7, 0.8, 0.0, 2.0 ];
var y = [ 1.9, 0.8, 1.1, 0.1, -0.1, 4.4, 5.5, 1.6, 4.6, 3.4 ];

var out = pcorrtest( x, y, {
    'alternative': 'less'
});
var table = out.print();
/* e.g., returns
    t-test for Pearson correlation coefficient

    Alternative hypothesis: True correlation coefficient is less than 0

        pValue: 0.997
        statistic: 3.709
        95% confidence interval: [-1,0.9363]

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

out = pcorrtest( x, y, {
    'alternative': 'greater'
});
table = out.print();
/* e.g., returns
    t-test for Pearson correlation coefficient

    Alternative hypothesis: True correlation coefficient is greater than 0

        pValue: 0.003
        statistic: 3.709
        95% confidence interval: [0.433,1]

    Test Decision: Reject null in favor of alternative at 5% significance level
*/

To test whether the correlation coefficient is equal to some other value than 0, set the rho option. Hypotheses tests for correlation coefficients besides zero are carried out using the Fisher z-transformation.

var x = [ 0.7, -1.6, -0.2, -1.2, -0.1, 3.4, 3.7, 0.8, 0.0, 2.0 ];
var y = [ 1.9, 0.8, 1.1, 0.1, -0.1, 4.4, 5.5, 1.6, 4.6, 3.4 ];

var out = pcorrtest( x, y, {
    'rho': 0.8
});
/* e.g., returns
    {
        'alpha': 0.05,
        'rejected': false,
        'pValue': ~0.972,
        'statistic': ~-0.035,
        'ci': [ ~0.332, ~0.949 ],
        'nullValue': 0.8,
        'pcorr': ~0.795,
        // ...
    }
*/

var table = out.print();
/* e.g., returns
    Fisher's z transform test for Pearson correlation coefficient

    Alternative hypothesis: True correlation coefficient is not equal to 0.8

        pValue: 0.972
        statistic: -0.0351
        95% confidence interval: [0.3315,0.9494]

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

Examples

var rnorm = require( '@stdlib/random-base-normal' );
var sqrt = require( '@stdlib/math-base-special-sqrt' );
var pcorrtest = require( '@stdlib/stats-pcorrtest' );

var table;
var out;
var rho;
var x;
var y;
var i;

rho = 0.5;
x = new Array( 300 );
y = new Array( 300 );
for ( i = 0; i < 300; i++ ) {
    x[ i ] = rnorm( 0.0, 1.0 );
    y[ i ] = ( rho * x[ i ] ) + rnorm( 0.0, sqrt( 1.0 - (rho*rho) ) );
}

out = pcorrtest( x, y );
table = out.print();
console.log( table );

out = pcorrtest( x, y, {
    'rho': 0.5
});
table = out.print();
console.log( table );

Notice

This package is part of stdlib, a standard library for JavaScript and Node.js, with an emphasis on numerical and scientific computing. The library provides a collection of robust, high performance libraries for mathematics, statistics, streams, utilities, and more.

For more information on the project, filing bug reports and feature requests, and guidance on how to develop stdlib, see the main project repository.

Community

License

See LICENSE.

Copyright

Copyright © 2016-2024. The Stdlib Authors.