In short

Fastest solution:

let a = new Array(n); for (let i=0; i<n; ++i) a[i] = 0;

Shortest (handy) solution (3x slower for small arrays, slightly slower for big (slowest on Firefox))

Array(n).fill(0)

Details

Today 2020.06.09 I perform tests on macOS High Sierra 10.13.6 on browsers Chrome 83.0, Firefox 77.0, and Safari 13.1. I test chosen solutions for two test cases

• small array - with 10 elements - you can perform test HERE
• big arrays - with 1M elements - you can perform test HERE

Conclusions

• solution based on new Array(n)+for (N) is fastest solution for small arrays and big arrays (except Chrome but still very fast there) and it is recommended as fast cross-browser solution
• solution based on new Float32Array(n) (I) returns non typical array (e.g. you cannot call push(..) on it) so I not compare its results with other solutions - however this solution is about 10-20x faster than other solutions for big arrays on all browsers
• solutions based on for (L,M,N,O) are fast for small arrays
• solutions based on fill (B,C) are fast on Chrome and Safari but surprisingly slowest on Firefox for big arrays. They are medium fast for small arrays
• solution based on Array.apply (P) throws error for big arrays

  function P(n) {
    return Array.apply(null, Array(n)).map(Number.prototype.valueOf,0);
  }
  
  try {
    P(1000000);
  } catch(e) { 
    console.error(e.message);
  }

Code and example

Below code presents solutions used in measurements

function A(n) {
  return [...new Array(n)].fill(0);
}

function B(n) {
  return new Array(n).fill(0);
}

function C(n) {
  return Array(n).fill(0);
}

function D(n) {
  return Array.from({length: n}, () => 0);
}

function E(n) {
  return [...new Array(n)].map(x => 0);
}

// arrays with type

function F(n) {
  return Array.from(new Int32Array(n));
}

function G(n) {
  return Array.from(new Float32Array(n));
}

function H(n) {
  return Array.from(new Float64Array(n)); // needs 2x more memory than float32
}

function I(n) {
  return new Float32Array(n); // this is not typical array
}

function J(n) {
  return [].slice.apply(new Float32Array(n));
}

// Based on for

function K(n) {
  let a = [];
  a.length = n;
  let i = 0;
  while (i < n) {
    a[i] = 0;
    i++;
  }
  return a;
}

function L(n) {
  let a=[]; for(let i=0; i<n; i++) a[i]=0;
  return a;
}

function M(n) {
  let a=[]; for(let i=0; i<n; i++) a.push(0);
  return a;
}

function N(n) {
  let a = new Array(n); for (let i=0; i<n; ++i) a[i] = 0;
  return a;
}

function O(n) {
  let a = new Array(n); for (let i=n; i--;) a[i] = 0;
  return a;
}

// other

function P(n) {
  return Array.apply(null, Array(n)).map(Number.prototype.valueOf,0);
}

function Q(n) {
  return "0".repeat( n ).split("").map( parseFloat );
}

function R(n) {
  return new Array(n+1).join('0').split('').map(parseFloat)
}

// ---------
// TEST
// ---------
[A,B,C,D,E,F,G,H,I,J,K,L,M,N,O,P,Q,R].forEach(f => {
  let a = f(10); 
  console.log(`${f.name} length=${a.length}, arr[0]=${a[0]}, arr[9]=${a[9]}`)
});

This snippets only present used codes

Example results for Chrome:

function zeroFilledArray(size) {
    return new Array(size + 1).join('0').split('');
}

To create an all new Array

new Array(arrayLength).fill(0);

To add some values at the end of an existing Array

[...existingArray, ...new Array(numberOfElementsToAdd).fill(0)]

Example

//**To create an all new Array**

console.log(new Array(5).fill(0));

//**To add some values at the end of an existing Array**

let existingArray = [1,2,3]

console.log([...existingArray, ...new Array(5).fill(0)]);

The fastest here is

(arr = []).length = len; arr.fill(0);

const item = 0
const arr = Array.from({length: 10}, () => item)

const item = 0
const arr = Array.from({length: 42}, () => item)
console.log('arr', arr)

let arr = [...Array(100).fill(0)]

You can check if index exist or not exist, in order to append +1 to it.

this way you don't need a zeros filled array.

EXAMPLE:

var current_year = new Date().getFullYear();
var ages_array = new Array();

for (var i in data) {
    if(data[i]['BirthDate'] != null && data[i]['BirthDate'] != '0000-00-00'){

        var birth = new Date(data[i]['BirthDate']);
        var birth_year = birth.getFullYear();
        var age = current_year - birth_year;

        if(ages_array[age] == null){
            ages_array[age] = 1;
        }else{
            ages_array[age] += 1;
        }

    }
}

console.log(ages_array);

In my testing by far this is the fastest in my pc

takes around 350ms for 100 million elements.

"0".repeat(100000000).split('');

for the same number of elements map(()=>0) takes around 7000 ms and thats a humungous difference

I know this isn't the purpose of the question, but here's an out-of-the-box idea. Why? My CSci professor noted that there is nothing magical about 'cleansing' data with zero. So the most efficient way is to NOT do it at all! Only do it if the data needs to be zero as an initial condition (as for certain summations) -- which is usually NOT the case for most applications.