Weighted arithmetic mean (average) in JavaScript

weightedMean.js

function weightedMean(arrValues, arrWeights) {

  var result = arrValues.map(function (value, i) {

    var weight = arrWeights[i];
    var sum = value * weight;

    return [sum, weight];
  }).reduce(function (p, c) {

    return [p[0] + c[0], p[1] + c[1]];
  }, [0, 0]);

  return result[0] / result[1];
}

weightedMean([251, 360, 210], [0.1, 0.5, 0.7]);
// => 270.8461538461539

I implemented it this way:
https://gist.github.com/afuggini/ccc74896676751b50f471ec571ea9f5d

const sumArrayValues = (values) => {
  return values.reduce((p, c) => p + c, 0)
}

const weightedMean = (factorsArray, weightsArray) => {
  return sumArrayValues(factorsArray.map((factor, index) => factor * weightsArray[index])) / sumArrayValues(weightsArray)
}

weightedMean([251, 360, 210], [0.1, 0.5, 0.7]);
// => 270.8461538461539

Thank you very much!

Here is the TypeScript / ES6 variation for people who are allergic to var and function:

export const weightedMean = (arrValues: number[], arrWeights: number[]) => {
  const result = arrValues
    .map((value, i) => {
      const weight = arrWeights[i]
      const sum = value * weight
      return [sum, weight]
    })
    .reduce((p, c) => [p[0] + c[0], p[1] + c[1]], [0, 0])

  return result[0] / result[1]
}

@verekia Nice work! I'll use that one in the future 👍

const tuples = [
    [100, 1],
    [200, 2],
    [300, 3]
];

const [valueSum, weightSum] = tuples.reduce(([valueSum, weightSum], [value, weight]) =>
    ([valueSum + value * weight, weightSum + weight]), [0, 0]);

console.log(valueSum / weightSum); // 233.333..4

Here is another way to do it. This is subjective but I like to break out variables because I find it easier to read.:

const tuples = [
    [100, 1],
    [200, 2],
    [300, 3]
]
const weightSum = tuples.reduce((accumulator, item) => accumulator + item[1], 0)
const weightedAverage = tuples.reduce((accumulator, item) => accumulator + item[0] * item[1] / weightSum, 0)

@franceindia

I very much like this approach, it is perfect for my use case. Thank you for sharing.