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
- :
- Re: Variance Components

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
- Mute
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Sep 19, 2016 3:44 AM
(8739 views)

Hi users!

I would like to obtain the variance components of a process giving a numerical result, including operators, instruments, lots and days I think there are two ways of analyzing this

-option "Fit model" and define the model

-option "Variability Gauge" and define the groupings and the Sample ID

Can anyone explain to me what the difference is between both modules and when to use one or the other?

Thanks for your help.

Best,

Dave

2 ACCEPTED SOLUTIONS

Accepted Solutions

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

David,

To me the biggest difference is that you can use fit model to look for cause and effect, and comparison and along with that you get a model to predict what will happen at various "what if " settings. The prediction profiler is a great tool for just that purpose. With the variability gauge you can look for cause and effect, but you cannot predict you can only compare. In both cases you should be able to find the factor that is contributing most to the variation you are seeing.

HTH

Bill

Highlighted

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

I agree with Bill. I use var comp for **visualization** and to see how the variances partition (where is the variability?). You can do a GAGE R&R study as well.

Fit model gives you stat. significance and prediction ability, etc. Both are very useful depending on your dataset...

3 REPLIES 3

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

Re: Variance Components

Please can anyone shed some light on my doubt? Any links or literature?

Thanks!

David

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

David,

To me the biggest difference is that you can use fit model to look for cause and effect, and comparison and along with that you get a model to predict what will happen at various "what if " settings. The prediction profiler is a great tool for just that purpose. With the variability gauge you can look for cause and effect, but you cannot predict you can only compare. In both cases you should be able to find the factor that is contributing most to the variation you are seeing.

HTH

Bill

Highlighted

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

I agree with Bill. I use var comp for **visualization** and to see how the variances partition (where is the variability?). You can do a GAGE R&R study as well.

Fit model gives you stat. significance and prediction ability, etc. Both are very useful depending on your dataset...