Probability density function

The Probability Density Function (PDF) gives information about how likely a value of a random variable is. It is not directly the probability but the probability per unit length. For a given random variable X we define the PDF px(x) as follows:

pX(x)=lim0 P(x<X<x+) =lim0=FX(x+) - FX(x)
pX(x)=dFX(x)dx

For step discontinuities in the CDF, the Dirac delta function is used to describe the PDF.

Properties:

  1. P(xx0)=FX(x0)=-x0pX(x)dx
  2. pX(x)0, x
  3. -+pX(x)dx=1
  4. P(a<x<b)=abpx(x)dx

The PDF of discrete random variables accumulate the probability in a finite and numerable number of values, which are the possible values. Its PDF can be expressed using the Dirac delta function:

pX(x)=  i-1N P(X=xi)δ(x-xi)
Project

  • University of Boras logo
  • UHI logo
  • Alcala University logo
  • Digital connextions logo

This resource was developed as part of an Erasmus+ project, funded with support from the European Commission under grant agreement 2016-1-SE01-KA203-22064.

The project was a collaboration between:

  Creative Commons License

This resource has been released under Creative Commons license CC-BY-SA 4.0.

Contact

  • University of Boras logo
  • UHI logo
  • Alcala University logo
  • Digital connextions logo

If you would like more information on this resource please contact:

  • Academic content – The University of Alcalá (www.uah.es/en/)
  • Technical resource development – The University of the Highlands and Islands Educational Development Unit - EDU (edu@uhi.ac.uk)
Disclaimer

  • University of Boras logo
  • UHI logo
  • Alcala University logo
  • Digital connextions logo

Except where otherwise noted, this website is licensed under Creative Commons license CC-BY-SA 4.0. All images used under permission remain the copyright of the license holder.

PDF

  • University of Boras logo
  • UHI logo
  • Alcala University logo
  • Digital connextions logo

Download a copy of this resource in PDF format.

You can also print individual pages by printing directly from the browser.

×