PDFs, CDFs, and other ‘Fs’
What the hell are they?
Speaker: Chris Jackson
Ok. They say there are lies, damn lies and statistics. And at the heart of statistics are these intimidating functions like ‘probability density functions’ and ‘cumulative distribution functions.’
They mean something – but are practically useless unless they help inform a decision. And we as engineers are supposed to know what these functions mean. But have you ever sat in a meeting or presentation where you realized there were some gaps in your knowledge? … or perhaps you felt like you were given too many charts but not enough help to make your decision?
If you want a really easy introduction or review of these functions – then check out this webinar!
This Accendo Reliability webinar originally broadcast on 28 April 2020.
The audio track is now an Accendo Reliability Webinar Series podcast episode. View the episode show notes to listen or subscribe to podcast
Please login with your site registration to view the video recording.
If you haven’t registered, it’s free and takes only a moment.
Really nice webinar, Chris. Finally I feel comfortable with failure rate, failure function and how these are derived.
Further topics:
1.) Confidence intervals for CDF/PDF/failure functions and use in RE
2.) Preventive maintenance versus failure rate
3.) CDF / failure rates and uses with predictive / prescriptive maintenance
Thanks!!
So true!!!
Evan … I missed this comment notification – so sorry for not responding sooner! Firstly – I appreciate the feedback.
In regard to the proposed webinars … they are really good! I will add the first one to the list. The second one I believe I have since covered in another webinar – could you confirm if you have seen it? And … I might need a little more guidance for the third topic. Typically predictive maintenance refers to CBM … which is based on sensors and warnings versus failure rates.
Is there something I am missing?
Thanks again for the comments.
What are the units of the Y axis for the histogram discussed initially?
I understand that for a pdf the Y axis is (1/x axis units).
Hi Mariraja,
Same apology as for Evan as I missed this notification. Sorry!
The units for the Y-axis of the histogram which appears at the 2:50 mark is simply the number of data points within the width of each column. So the height of each bar helps us estimate the density of data points, which will eventually approximate the PDF (bell curve) if we get lots and lots of them.
Hope that helps!
Chris