univariate-tests

Univariate tests to detect outlier.

Tests available

• Grubbs
• Dixon

Installation

$ npm i univariate-tests

Usage

Grubbs

import { grubbs } from 'univariate-tests';
const testValues = [10.45, 10.26, 10.49, 10.36, 10.53, 10.77];
const { test, criticalValue } = grubbs(testValues);
console.log(criticalValue); // 1.82
console.log(test[0]); // { value: 10.45, score: 0.15378928962176208, pass: true }

Dixon

import { dixon } from 'univariate-tests';
const testValues = [10.45, 10.26, 10.49, 10.36, 10.53, 10.77];
const { test, criticalValue } = dixon(testValues);
console.log(criticalValue); // 0.625
console.log(test[0]); // { value: 10.26, score: 0.1960784313725484, pass: true }

References

• Miller, J. N., & Miller, J. C. (2010). Statistics and Chemometrics for Analytical Chemistry.

License

MIT