Here, we are going to make the task more complicated; let's talk about Poisson distribution.

To work with this distribution, we should import the poisson object from scipy.stats, and we you can apply numerous functions to this distribution like pmf, sf, and cdf that were already learned.

Key characteristics:

It measures the frequency over a specific time interval.

Example:

The data on how often your application had a certain number of users during its entire existence.

Explanation:

It seems to me that you know a lot now, so in this chapter, we are going to create and explain the Poisson distribution. Here is the code:

We are going to talk about it a little bit, you already know the .rvs() function, but let's clarify something for this case poisson.rvs(mu = 50, size = 10000):

• mu means mean (here, it was defined randomly).
• size is the number inverted by the sum of all values ​​of the frequencies of the columns

Let's recall some functions, but for Poisson distribution (they are a little bit different):

For calculating the probability of receiving exactly k events: norm.pmf(k, mu).

For calculating the probability of receiving k or more events: norm.sf(k, mu).

For calculating the probability of receiving k or less events: norm.cdf(k, mu).

• mu is the mean value of the distribution.

Note

So, the probability that your app will visit an extremely small or large amount of people is extremely small. By the way, if the probability is too small, you can just drop it :)