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

Showing results for

- JMP User Community
- :
- Discussions
- :
- How to select independent variable in mixed model ...

- 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

Apr 5, 2017 7:11 AM
(706 views)

Hello,

I'm trying to run a mixed model in jmp 13 (not jmp pro) and need to add an independent variable. Does anyone know how to do this after selecting a "Fit Model" option?

Thank you in advance.

1 ACCEPTED SOLUTION

Accepted Solutions

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

Apr 6, 2017 7:23 AM
(1244 views)

Solution

The independent variable (or predictor, feature, or factor) is represented by a data column in your data table. Try this:

- Select
**Analyze**>**Fit Model**. - Select the column for the
**independent**variable. - Click
**Add**. (Note that you can make any linear predictor that you might need using Add, Cross, or one of the Macros with the Degree value.) - If this term is meant to be a random effect, then select it in the Effects list, click the red triangle next to Attributes, and select
**Random**.

This sequence will force JMP to use REML to estimate the parameters including the variance components.

Learn it once, use it forever!

3 REPLIES

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

Apr 6, 2017 5:12 AM
(669 views)

I'm sorry, but I'm not sure that I have understood your specific question. Perhaps this helps.

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

Apr 6, 2017 7:23 AM
(1245 views)

The independent variable (or predictor, feature, or factor) is represented by a data column in your data table. Try this:

- Select
**Analyze**>**Fit Model**. - Select the column for the
**independent**variable. - Click
**Add**. (Note that you can make any linear predictor that you might need using Add, Cross, or one of the Macros with the Degree value.) - If this term is meant to be a random effect, then select it in the Effects list, click the red triangle next to Attributes, and select
**Random**.

This sequence will force JMP to use REML to estimate the parameters including the variance components.

Learn it once, use it forever!

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

Apr 6, 2017 8:49 AM
(657 views)