lregression

1.0.0 • Public • Published

Simple Linear Regression

Very basic linear regression

Usage

The constructor has two optional arguments:

  • data An array of pairs in the form [x,y]
  • clearData A boolean value (by default true) that determines if the data array will be cleared after analyzing the data to save memory.

The library exposes two methods:

  • analize Used to calculate correlation, slope, mean, standard deviation and function intercept.
  • predict(x) Calculates the value of y for a value x

Initialization with data

const LR = require('lregression');
const regression = new LR();
const data = [ [1,2], [2,4], [3,6] ];

console.dir(regression.analize());

regression.analize();
for (let i = 4; i < 10; i++) {
	console.log(`x=${i} y=${regression.predict(i)}`);
}

Manually feeding the data

const LR = require('lregression');
const regression = new LR();

for (let i = 0; i < 50; i++) {
	regression.add(i, i * 2);
}

console.dir(regression.analize());

regression.analize();
for (let i = 51; i < 100; i++) {
	console.log(`x=${i} y=${regression.predict(i)}`);
}

Readme

Keywords

Package Sidebar

Install

npm i lregression

Weekly Downloads

1

Version

1.0.0

License

MIT

Unpacked Size

4.21 kB

Total Files

5

Last publish

Collaborators

  • elfido