+- 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: Probability Wizard (/thread-16696.html)
Probability Wizard - Wes Loewer - 04-19-2021 07:47 PM
In Tim's announcement, the first item listed in the release notes is
Quote:New Functionality and Changes (Calculator Software)
Probability wizard
About a year or so ago, there was a thread here about the need for a Probability Wizard. The collection of Wizards that I posted in that thread wasn't bad considering that it was written within the confines of PPL, but the new Probability Wizard is a much slicker tool in every respect.
For those of you who are interested in Stats, check it out. It's found in the Toolbox, Math, Probability, Prob. Wizard...
Take note of the Linked option in the soft menu. It allows you to indicate whether or not you want the left and right bounds to be linked together such that there are equal areas to the left and right of the bounds. This is a very useful feature for finding Critical Values. (But don't forget to turn Linked off when you don't want it or you will get very unexpected results.)
It looks like the visual representations have issues as the incorrect areas are shaded, but otherwise the Wizard looks great.
RE: Probability Wizard - Wes Loewer - 04-22-2021 07:31 PM
The Probability Wizard is very good, but there are a few small tweaks that would really help.
- The Linked option is exceedingly useful, however I recommend that it be turned off by default. Let the user turn it on when they need it. I've been confused numerous times about the results until I realized that I had forgotten that the the bounds are linked by default.
- It would be helpful if the values and settings were persistent between runs. Often you want to use the same distribution with the same type and some of the same values. It would be nice not to have to repeat resetting the options and re-entering the values each time.
- The displayed graphs are a nice touch for visual feedback, but the Normal and Chi Squared distributions do not shade the correct regions.