Programming the statistical procedures from SAS

Clustering

Accepted Solution Solved
Reply
Contributor
Posts: 70
Accepted Solution

Clustering

Hello

 

1) I am getting warning about correlated variables as shown in the screenshot, is there any specific reason for this?.

 

2) How to profile and interpret clusters when an interaction variable is used.

warning.PNG

Thanks


Accepted Solutions
Solution
‎06-16-2017 06:47 AM
PROC Star
PROC Star
Posts: 188

Re: Clustering

What do I do when confronted with a mysterious warning or error? Search on the text.

 

Googling "warning: the two values above are invalid for correlated variables" yields:

 

http://support.sas.com/kb/45/440.html

and

https://support.sas.com/documentation/cdl/en/statug/63347/HTML/default/viewer.htm#statug_fastclus_se...

 

among others. The internet can be a rich resource, use it.

 

View solution in original post


All Replies
Solution
‎06-16-2017 06:47 AM
PROC Star
PROC Star
Posts: 188

Re: Clustering

What do I do when confronted with a mysterious warning or error? Search on the text.

 

Googling "warning: the two values above are invalid for correlated variables" yields:

 

http://support.sas.com/kb/45/440.html

and

https://support.sas.com/documentation/cdl/en/statug/63347/HTML/default/viewer.htm#statug_fastclus_se...

 

among others. The internet can be a rich resource, use it.

 

Contributor
Posts: 70

Re: Clustering

Thank you.

 

How an analyst can decide which of the following method in Heirarchical clustering to be used in which scenario.

 

Single Linkage, Complete linkage, centroid linkage.

 

 

PROC Star
PROC Star
Posts: 188

Re: Clustering

This thread from Cross Validated addresses your question--and notes that it's a complicated topic:

 

https://stats.stackexchange.com/questions/195446/choosing-the-right-linkage-method-for-hierarchical-...

 

It could be a good place to start.

☑ This topic is solved.

Need further help from the community? Please ask a new question.

Discussion stats
  • 3 replies
  • 159 views
  • 1 like
  • 2 in conversation