I am running a sequential GMM with 3 clusters of change on 2 variables of interest. My sample size isn't huge (aprox 2000).
Some of the patterns I get are inconsequential - i.e. people who are in cluster 3 on variable x don't tend to also be in cluster 3 on variable y (n of 7). I would like to know if there is a way of suppressing patterns with small n's? Obviously I have 9 different patterns, only some of which I believe are theoretically interesting to include in additional research (I would like to examine differences in psychological outcomes based on pattern category).