EN
JavaScript - moving average
11
points
In this article, we would like to show how to calculate the moving average in JavaScript
Quick solution:
// ONLINE-RUNNER:browser;
const calculateItemsSum = (data, start, stop) => {
let sum = 0;
for (let j = start; j < stop; ++j) {
sum += data[j];
}
return sum;
};
const caculateMovingAverage = (data, window) => {
const steps = data.length - window;
const result = [ ];
for (let i = 0; i < steps; ++i) {
const sum = calculateItemsSum(data, i, i + window);
result.push(sum / window);
}
return result;
};
// Usage example:
const data = [1, 2, 3, 4, 5, 6];
const average = caculateMovingAverage(data, 3); // where 3 is a window size
console.log(average); // [2, 3, 4]
Where:
windowmeans a number of elements used to calculate the average during one step.
Note: the above solution has O(n^2) complexity.
Optimal solution example
The presented solution in this section has only O(n) complexity.
// ONLINE-RUNNER:browser;
function caculateMovingAverage(data, window) {
var result = [ ];
if (data.length < window) {
return result;
}
var sum = 0;
for (var i = 0; i < window; ++i) {
sum += data[i];
}
result.push(sum / window);
var steps = data.length - window - 1;
for (var i = 0; i < steps; ++i) {
sum -= data[i];
sum += data[i + window];
result.push(sum / window);
}
return result;
}
// Usage example:
var data = [1, 2, 3, 4, 5, 6];
var average = caculateMovingAverage(data, 3); // where 3 is a window size
console.log(average); // [2, 3, 4]
Warning: the above solution uses
-=and+=operations on thesumvariable that makes numerical mistakes when we work on floating numbers - it is good to modify below source code and recomputesumonce per some number of iterations (number may be selected empiricaly).