PROC MODECLUS is applied to a multivariate dataset. What type of clustering algorithm does PROC MODECLUS implement?
Select an answer to reveal the explanation.
Short Explanation
Here's the deal — PROC MODECLUS uses nonparametric mode-seeking, density-based clustering. It estimates density with kernel density estimation, finds local density peaks, and assigns points to nearby modes. You get clusters as high-density regions separated by lower-density boundaries.
Full Explanation
The density-based clustering that identifies clusters as high-density regions separated by lower-density boundaries is correct because PROC MODECLUS implements nonparametric mode-seeking clustering. It estimates the density at each point using kernel density estimation and identifies clusters as modes, or local density peaks, with observations assigned to nearby modes. This approach can find clusters of arbitrary shape and automatically determines the cluster count. The K-means partitional clustering with a fixed number of clusters is PROC FASTCLUS or PROC HPCLUS. The hierarchical agglomerative clustering with Ward's minimum variance criterion is PROC CLUSTER with METHOD=WARD. The model-based clustering using Gaussian mixture models with EM parameter estimation is not a standard SAS base procedure and is available in SAS Enterprise Miner.