cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Choose Language Hide Translation Bar
GaneshJhawar
Level I

Seeking help on discrete choice modelling with jmp

Dear All,

 

I am currently creating a model for my masters and i am new to jmp world. I have collected all the data from surveys but now confused if it makes sense to use choice design or maxdiff design for this kind of analysis. Hope the wide audience present here could help me. 

 

Here is a brief of my data points: 

A procurement manager whose responsibility is to select containers for his shipments on a daily basis. He has a set of options and he picks 1 out of all the options. For all the options there are different terms which the manager takes into account before selecting the containers. 

For example: Let X be the Procurement manager. He has options:

A can supplies (1 container)(Price $500 each)(From- China to America)(60 days shipment)($2.5 storage charges)($5000 container price)(basic insurance)

B can supply (2 container)(Price $600 each)(From- China to America)(50 days shipment)($3 storage charges)($4000 container price)(premium insurance) etc. 

He selected option A.

So, like this i will be having several data points where i have collected information which explains which data is selected and which data has been declined. 

Now I would like to know how can I use this in JMP. Can anyone please help me

 

1 REPLY 1
peng_liu
Staff

Re: Seeking help on discrete choice modelling with jmp

I think the decision should have been made before you hand out survey. Or you won't have desired data.

If you have already collected data, one technical difference between the two is that Choice model has a single indicator response (pick one out of a set) and MaxDiff has two-indicator response (pick two out of a set, best and worst.)

References:

https://www.jmp.com/support/help/en/17.1/?os=win&source=application#page/jmp/choice-models.shtml#

https://www.jmp.com/support/help/en/17.1/?os=win&source=application#page/jmp/maxdiff.shtml#