cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Browse apps to extend the software in the new JMP Marketplace
Choose Language Hide Translation Bar
ask
ask
Level I

What is the best way to use JMP to do clustering analysis if the data is a mixed model

What is the best way to use JMP to do clustering analysis if the data is a mixed model with continuous and categorical variables? K-Means seems to just work for continuous variables. Right? Thanks in advance.

1 REPLY 1
Victor_G
Super User

Re: What is the best way to use JMP to do clustering analysis if the data is a mixed model

Hi @ask,

 

K-Means and Normal Mixtures are clustering methods requiring numeric columns only. You can have more details and a comparison between clustering methods here : Overview of Platforms for Clustering Observations

 

If you have a dataset with categorical and numeric columns, you can try the Hierarchical Cluster platform.

In this platform, you have access to various methods (Ward, Average, Centroïd, ...) , and since Hierarchical clustering do not require to specify the number of clusters apriori, this approach is interesting as it can provide an exhaustive overview of the possible links between your observations/samples at different levels.

More infos available here : Clustering | JMP

 

I hope this answer will help you, 

Victor GUILLER
L'Oréal Data & Analytics

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