# How can we calculate probability density of a histogram?

218 views (last 30 days)
ishita agrawal on 19 Aug 2017
Commented: Image Analyst on 28 Jun 2022
Hello, I have a dataset containing hundreds of data points representing duration of events. I have plotted a histogram for this data set. I want to calculate probability density for the same. I want a plot having event duration at X-axis and probability density at Y-axis. I have attached histogram for which I want to calculate PD.

the cyclist on 19 Aug 2017
Edited: the cyclist on 19 Aug 2017

If you have version R2014b or later, you can use the histogram command, and specify the 'Normalization' property to be 'probability':

```rng default
x = randn(1000,1);
h = histogram(x,'Normalization','probability');
```

h.Values will give you the density.

ishita agrawal on 19 Aug 2017
It worked. Thank you

### More Answers (3)

Image Analyst on 19 Aug 2017
Did you look in the documentation for histogram and see the normlization option:
'Normalization' Type of normalization
'count' (default) | 'probability' | 'countdensity' | 'pdf' | 'cumcount' | 'cdf'
##### 3 CommentsShow 1 older commentHide 1 older comment
Sajid Afaque on 12 Jun 2020
what about for prior version.
i am using 2013b and i need probability distribution
Image Analyst on 12 Jun 2020
Just divide your counts by either the number of counts in the highest bin, or by the sum of all bins, depending on how you define normalize.

Steven Lord on 19 Aug 2017
Use the 'Normalization' option of the histogram function.

Mehri Mehrnia on 27 Jun 2022
I have the same problem , just at I want to have same edges in both before normalization and after normalization.
I got an strange result!!!
Image Analyst on 28 Jun 2022
Thanks for the announcement. If you have any questions, then attach your data and code to read it in with the paperclip icon after you read this: