0
$\begingroup$

first of all, thank you for your time, here is my question;

Is it possible to find emprical PDF by using normalized histogram?

I am trying to learn discrete event simulation and what I see is there is lots of statistical models which I am not so much familiar with. More specificly;

I have data points such as ; 26.5 74.1 80.3 72.3 87.7 60.9 98.3 52.1 40.8 74.3 13.4 92.5 79.2 74.1 84.1 42.6 57.3 65.4 66.5 72.5 55.2 33.2 79.2 33.9 18.4 83.6 10.4 56.3 82.5 99.8 49.9 19.6 68.3 45.4 59.9 16.8

And I find histogram like this

Histogram

From there, what I really want is find the emprical PDF by using histogram that I obtained and make a graph like this;

enter image description here

Please note that, this is not a homework or something like that and also I actually solving and constructing histogram by hand. After that When I find this PDF, I will compare mean and variance by using it. I am assuming that I have to look to the histogram and make an assumption maybe?

Thank you again

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

If you standardize your histogram, you get a probability density function. In your case, this pdf will be a very poor approximation of the real pdf that rendered the process you measured, since you got only 36 data points. Having so few data, I would only compute the median and the mean of the data. Since your data are numerical, I wouldn't fit any model with less than 100 data points.

Improve this answer
$\endgroup$
1
  • $\begingroup$ Thank you for your contribution. I know data is not enough but I was just warming myself on concept. $\endgroup$
    – Ozzzy
    Apr 11, 2019 at 14:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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