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

Showing results for

- JMP User Community
- :
- Discussions
- :
- Data Analysis

- 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

Oct 21, 2015 12:17 PM
(483 views)

I have a set of data (A = 1X32) and another set of data (B = 1X32) and I have a third column which is (C) C=(B-A)/A. Each element in the matrix has its own name what I am trying to get is that the highest 10 values of (C) with their names. That can be done manually by using excel but I want to use more sophisticated way to do that.

Is that possible in JMP12?

2 REPLIES

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

Oct 21, 2015 4:02 PM
(273 views)

Hi, asaalsaadi10.

Of course, it's possible. But don't use JMP because it's sophisticated. Use JMP because it does operations and math correctly (unlike Excel).

How would you do it in Excel? Sort C descending, and pick the first 10 rows? That's easily done in JMP or JSL. But some questions come to mind.

What if 12 (for instance) values of C tie at the maximum? Do you want 10 and only 10 rows chosen arbitrarily? Or do you want the names of all 12?

Will you ever do this for datasets much larger? What if all the As and Bs are missing?

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

Oct 22, 2015 8:11 AM
(273 views)

Hi Kevin,

The problem that I am facing at the moment is the names of the max 10 values (this is the most important element at the moment), these values are labelled with specific names. If I end up with 12 values have that will be fine I can then eliminate some of them depending on our calculations.