HP Forums
Normal Distribution - Printable Version

+- HP Forums (https://www.hpmuseum.org/forum)
+-- Forum: HP Calculators (and very old HP Computers) (/forum-3.html)
+--- Forum: HP Prime (/forum-5.html)
+--- Thread: Normal Distribution (/thread-12463.html)



Normal Distribution - JDW - 02-19-2019 12:24 AM

I high school student posted a question under my YouTube video on the Prime yesterday, saying she is trying to find the best way to calculate "Normal Distribution." Today she said, "I found a slow but effective way to calculate normal distribution graphically, you can use NORMALD(mean, std deviation,x) in Function Symbolic View and use signed area for the upper and lower limits in plot mode. Also i read that you can use NORMAL_CDF(mean, std deviation, lower, upper) but this way is only effective symbolically."

What is the best way on the Prime to calculate Normal Distribution?


RE: Normal Distribution - mark4flies - 02-19-2019 01:38 PM

Yes, the CDF is the definite integral of the PDF for all distribution models.


RE: Normal Distribution - JDW - 02-19-2019 11:55 PM

But note that she said this:

“You can use NORMAL_CDF... but it is only effective symbolically.”

That statement implies she is looking for a non-symbolic solution. What thoughts do you have on that?

Thank you.


RE: Normal Distribution - Tim Wessman - 02-20-2019 01:34 PM

She is probably in the CAS view and using whole numbers only...

These are primarily numeric commands and the fact that she is getting symbolic results is by FAR the most unlikely case since you never have whole, exact numbers in real use cases.

Home screen, use ANY of the _CDF commands and you get numeric results.

NORMALD_CDF(0,1,0) --> .5

or in the CAS screen, add a decimal after any number to make it "approximate" and you get an immediate numeric result.


NORMALD_CDF(0.,1,0) ---> .5

Also, remind her that she can go to the MATH->Probability->Cumulative, select the one she wants and press help to get help on it with examples.


RE: Normal Distribution - Jlouis - 02-20-2019 04:11 PM

Thanks Tim, it was useful to me as well.

Cheers