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

- JMP User Community
- :
- Discussions
- :
- DOE output

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

Highlighted

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

Sep 21, 2018 3:11 PM
(5435 views)

Ran my DOE analysis and got an output shown below:

QTY{LowMCC&6D - High MCC}

Not sure how to decode that to understand which level I should use for the qty factor.

Thank you,

Rami

1 ACCEPTED SOLUTION

Accepted Solutions

Highlighted

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

not exactly. In the Stepwise platform, categorical variables (nominal and ordinal) are coded in a hierarchical fashion. This differs from coding in other least squares fitting platforms. In hierarchical coding, the levels of the categorical variable are successively split into groups of levels that most separate the means of the response. The splitting process achieves the goal of representing a k-level categorical variable by k - 1 terms.Here you find more about that: https://www.jmp.com/support/help/14/models-with-nominal-and-ordinal-effects.shtml

Pressing Next button on the above page you came to this page

As said before there is the option to use stepwise with whole effects to keep an easy interpretability. More you find in the rules section: https://www.jmp.com/support/help/14-1/stepwise-regression-control-panel.shtml#240750

The advantage of having the hierarchical way in stepwise is that you migth get certain group-level effects earlier into the model than using the whole effect. The price to pay is the interpretation capability for people who are not used to it.

Hope this helps .

Martin

4 REPLIES 4

Highlighted
##

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

Re: DOE output

Rami,

I believe more information is required to be able to help.

The Profiler should visually help you answer your question of what level you should set your QTY.

Highlighted
##

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

Re: DOE output

QTY{LowMCC&6D - High MCC} means that this is the effect between two groups: LowMCC and 6D on the one side - and Hig MCC on the other. As Lou pointed out in the profiler of the model you will see the effect in a change from one grouping to the other.Usually the scale will be from -1 to 1.If there is no other effect QTY{LowMCC - 6D} there might not be a significant effect between those two. But this is speculation on my side without having more information on the whole setup.

If you used stepwise you could use whole effects to have a better interpretation of the effects in the model and profiler.

Highlighted
##

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

Re: DOE output

Thank you for the response. Why would it be between two groups, when qty has three levels? Does this mean Low MCC and 6D behave the same?

Looking at the profiler, it states -1 to maximize output, however I dont know which level correspondes to -1.

Highlighted

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

not exactly. In the Stepwise platform, categorical variables (nominal and ordinal) are coded in a hierarchical fashion. This differs from coding in other least squares fitting platforms. In hierarchical coding, the levels of the categorical variable are successively split into groups of levels that most separate the means of the response. The splitting process achieves the goal of representing a k-level categorical variable by k - 1 terms.Here you find more about that: https://www.jmp.com/support/help/14/models-with-nominal-and-ordinal-effects.shtml

Pressing Next button on the above page you came to this page

As said before there is the option to use stepwise with whole effects to keep an easy interpretability. More you find in the rules section: https://www.jmp.com/support/help/14-1/stepwise-regression-control-panel.shtml#240750

The advantage of having the hierarchical way in stepwise is that you migth get certain group-level effects earlier into the model than using the whole effect. The price to pay is the interpretation capability for people who are not used to it.

Hope this helps .

Martin

Article Labels

There are no labels assigned to this post.