0.0.9 • Published 3 years ago
benfordslaw v0.0.9
benfordslaw
Check if a number array confirms to the Benford's Law
Heavily inspired by https://github.com/target/huntlib
Install
npm i benfordslawUsage
Look at the tests for good examples.
E.g.:
import { BenfordsLaw } from 'benfordslaw';
const numbers = [1,2,3,4,5,6,7,8,9];
const benfords = new BenfordsLaw(numbers);
const chiSquared = benfords.getChiSquared();
// chiSquared = 0.40105320411553363
const probability = benfords.getProbability()?.toFixed(1);
// probability = 1.0The ChiSquared result is a float and describes how well Benford's Law was matched. Lower is better.
The Probability describes how relevant ChiSquared is. It should be >= 0.9
getDist() returns the distribution of the numbers.
Why TypeScript
With Types TypeScript adds a extra layer of security. I also use mainly TypeScript projects with this project so TypeScript just makes sense. And TS also compiles to JavaScript Code so JS projects also can use it.
General Information
License: MIT
Author: Christoph Daniel Miksche
See also
- https://blog.m5e.de/post/benchmarking-go-crystal-python-and-js/ Benchmarking Benford's Law in different Programming Languages
- https://github.com/CMiksche/benford A Go Package implementation of this benford's law check
- https://github.com/CMiksche/benfordslaw.cr A Crystal implementation of this benford's law check