turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- JMP User Community
- :
- Discussions
- :
- Discussions
- :
- How to get out a formula for confidence region lines on Actual by Predicted Plot

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

May 1, 2014 12:36 PM
(1288 views)

Hello,

Can any one help me with finding the way to get out a formula for confidence region - red dash lines on Actual by Predicted Plot in Fit Model Report, so I can recreate the plot some where outside of JMP? Thanks!

3 REPLIES

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

May 9, 2014 11:22 AM
(1159 views)

The interval doesn't always correspond to a straightforward formula representation.You can export the graph as something like SVG or PDF which will provide curve components that can manipulated in graphics design software for post-production tuning.

[removed erroneous part about using Mean Confidence Intervals]

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

May 9, 2014 12:49 PM
(1159 views)

Thanks for your response! I tried to save 'Mean Confidence Interval' values before, but when I plot Lower Upper Mean vs my Predicted value, it gives me very scattered sets of points and it does not look the same as on JMP Actual vs Predicted plot red dot lines, please see attached picture. Should I fit them somehow to get lines?

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

May 9, 2014 1:18 PM
(1159 views)

Looks like my answer only holds for simple models, so I've removed that part of it. If someone with more stats knowledge doesn't chime in, you can try JMP tech support.