cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Try the Materials Informatics Toolkit, which is designed to easily handle SMILES data. This and other helpful add-ins are available in the JMP® Marketplace
Choose Language Hide Translation Bar
Thommy7571
Level I

Coding design versus numerical design

Hello,

would you consider it more reasonable to create a numerical design directly? Or first a coded design?

I first created a coded design (with .-1, 1 and 0).without randomisation.

is it possible to transform it afterwards in a numerical design by substituting the level values?

Is it possible to randomize it separately and to control the randomisation by a seed?

 

Sincerely

 

Thommy7571

1 ACCEPTED SOLUTION

Accepted Solutions
Victor_G
Super User

Re: Coding design versus numerical design

Hi @Thommy7571,

 

From my point of view, it is more useful to create the design with real values in JMP.

The column property "Coding" makes sure that the high value correspond to coded value +1 and low value to coded value -1 in the design, you don't need to do it manually afterwards.

 

If you already have a design with coded values, you can always edit the values using Recode Column Values.

You might also have to modify column property "Coding" so that the new high and low values are entered, and randomize the runs.

 

If you want to create reproducible designs (for example in the context of teaching), click on the red triangle during DoE creation on any DoE platform and click on "Set Random Seed" :

Victor_G_0-1719230061889.png

More infos on this option : Custom Design Options (jmp.com)

 

Hope this answer will help you,

Victor GUILLER

"It is not unusual for a well-designed experiment to analyze itself" (Box, Hunter and Hunter)

View solution in original post

1 REPLY 1
Victor_G
Super User

Re: Coding design versus numerical design

Hi @Thommy7571,

 

From my point of view, it is more useful to create the design with real values in JMP.

The column property "Coding" makes sure that the high value correspond to coded value +1 and low value to coded value -1 in the design, you don't need to do it manually afterwards.

 

If you already have a design with coded values, you can always edit the values using Recode Column Values.

You might also have to modify column property "Coding" so that the new high and low values are entered, and randomize the runs.

 

If you want to create reproducible designs (for example in the context of teaching), click on the red triangle during DoE creation on any DoE platform and click on "Set Random Seed" :

Victor_G_0-1719230061889.png

More infos on this option : Custom Design Options (jmp.com)

 

Hope this answer will help you,

Victor GUILLER

"It is not unusual for a well-designed experiment to analyze itself" (Box, Hunter and Hunter)