In this example, the data are sampled from two highly elongated multinormal distributions with equal covariance matrices. The following SAS statements produce Figure 8.16: data elongate; keep x y; ...
If the population clusters have very different covariance matrices, using PROC ACECLUS is of no avail. Although methods exist for estimating multinormal clusters with unequal covariance matrices ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results