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
- :
- Not all parameters calculated in Choice Model

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

Jan 21, 2017 3:36 PM
(3864 views)

I have followed the help for running a choice model on the attached Profile and Choice Data. I can't understand why the Tube, Generic, and None parameters don't show in the results (see attached JPG). There are four attributes, Format (4 levels), Shape (2 levels), Colour (3 levels) and Logo (2 levels). There are three profile ID choices, and the test is weighted by 'Test Cell' which seperates 4 different groups of people.

1 ACCEPTED SOLUTION

Accepted Solutions

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

The last level of each attribute is not shown in the parameter estimates table because it is not independent of the others. All of the parameter estimates for a particular attribute will sum to zero, so the last level will be the additive inverse of the sum of the other levels. You can choose Effect Marginals from the red triangle menu and you will see the values for all levels of each attribute.

4 REPLIES

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

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

Jan 23, 2017 12:03 PM
(3813 views)
| Posted in reply to message from susan_walsh1 01/23/2017 02:48 PM

@susan_walsh1THANK YOU! I did not click enough of the buttons to see that show up in JMP and have been trying to run an analysis in R instead for weeks. I am so grateful!

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

Is it also possible to display the std error for all parameters estimates including the marginal ones?

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

While I understand the degrees of freedom consideration, when running models that involve segmenting variables, it would be very handy to have the choice to see all the parameters in one place, including the ones not normally shown because they are redundant.