1
$\begingroup$

I have a website and I want to calculate the probability of clicks on the ads.

Let the probability that each user clicks on a link be p (something like 1%)

if we have totally N users, What is the formula that computes the probability of exactly n clicks? of course we have

0< = n <= N

Each users can click only once

$\endgroup$
0

1 Answer 1

3
$\begingroup$

Consider each user as a trial. For every trial you have two outcomes, they are success (clicks on ad) and failure (does not click on ad). $P[success] = p $ and $P[failure] = 1-p$.

The total number of ways in which $n$ users can be selected from $N$ users is $\binom{N}{n}$. So, the probability of exactly $n$ clicks is $\binom{N}{n}*p^n*(1-p)^m$ where $m = N-n$, as the clicks are independent and exactly $n$ clicks mean exactly $N-n$ 'not clicks'.

$\endgroup$
4
  • 1
    $\begingroup$ And this is called the binomial distribution. $\endgroup$
    – ken
    Oct 11, 2018 at 20:51
  • $\begingroup$ Yes @Ken sure it is!. $\endgroup$
    – naive
    Oct 12, 2018 at 14:35
  • $\begingroup$ Thanks @naive, is there any function to get (N n) for large Ns. (calculating N! for large Ns is impossible) an approximate function would be sufficient. $\endgroup$
    – Ormoz
    Oct 27, 2018 at 18:28
  • $\begingroup$ For large $N$ the binomial random variable can be described by a Normal distribution. $\endgroup$
    – naive
    Oct 28, 2018 at 7:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.