calculate the probability density function of Random variables

hi there, i am trying to calculate the PDF of Radomvariables usig Matlab. i generate some randomvariables (1000 Values,Normal distributed) and use them to Calculate the Output of a Powerflow Problem. the Calculated Output will also got 1000 Values now i want to calculate the PDF of my Output and Plot it (to see how my Output is distributed). my Problem is how can i calculate the PDF if i don't know how my funktion is distributed.

thanks a lot for your Help

Answer 1

[Image Analyst]

Use histcounts(). Divide by the number of observations to normalize.

[Image Analyst]

h g's "Answer" moved here:

Thanks for your Answer,

Why do i need to to normilize my Output? How can i define the bin ranges?

(The idea is: I want to observe how is my Output distributed if i use a Normaldistribution for my Input)

[Image Analyst]

You normalize it because histcounts() gives a count of the number of observations while the PDF is supposed to have an area of 1 by convention. So you need to divide the counts array by the total number of counts in the array to get an array that is normalized to an area of 1 and can be considered a PDF.

If you use bar() or plot() on your counts array or PDF array, and your distribution is normal, you should see the familiar normal Gaussian shape that you expect.

[h g]

i am using Matlab 2014a and there is no histcounts() function, can i use histc()? that means: if X is my Random Variable with 1000 numbers i have to write: bincounts=histc(X,1000)?? or how can i define the Binrange? thanks alot

[Walter Roberson]

You need to provide particular edges to histc, not the number of bins. hist() accepts the number of bins.