cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
JMP is taking Discovery online, April 16 and 18. Register today and join us for interactive sessions featuring popular presentation topics, networking, and discussions with the experts.
Choose Language Hide Translation Bar
david_arteta
Level III

Variance Components

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

Re: Variance Components

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

View solution in original post

Re: Variance Components

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...

View solution in original post

3 REPLIES 3
david_arteta
Level III

Re: Variance Components

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

Thanks!

David

Re: Variance Components

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

Re: Variance Components

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...