In SAS, a data scientist applies PROC HPCLUS to segment customers and wants to profile each cluster by computing mean values of key variables. Which post-clustering approach accomplishes this most efficiently?
Select an answer to reveal the explanation.
Short Explanation
You want PROC MEANS with a CLASS statement on the cluster output dataset. It quickly computes means by cluster assignment, which is exactly what you need to profile each cluster.
Full Explanation
After PROC HPCLUS writes cluster assignments to the output dataset, PROC MEANS with a CLASS statement on the cluster assignment variable, such as _CLUSTER_, is the efficient way to profile each cluster by computing means of key variables. It summarizes numeric variables by cluster without rerunning clustering. The PROC TABULATE approach grouped by _CLUSTER_ can summarize by cluster and provide formatted cross-tabulation output, but it is not the most efficient choice when the goal is simply means. The PROC FREQ approach with the cluster assignment variable in TABLES is also unsuitable because PROC FREQ produces frequency counts and categorical summaries, not means. Rerunning PROC HPCLUS with PROFILE=YES is wrong because that option does not exist.