I've created two different clusters, one using kmeans (partitioning method) and one using complete linkage (hierarchical).
–M is the mean number of individuals per cluster –SSW – Sum of squares within groups (from anova) –SST – total sum of squares (from anova).(Very easy to calculate in Stata).(Assumes equal sized groups, but it s close enough) SST SSW M M ICC u 1.
The clusters using the hierarchical method:
The clusters using the partitioning method:
This creates two different cluster. Now i want to visually see the clusters created. I want to do a scatter and make it show different color depending of the value of pcm_1 and c1. However, I can only create a scatter checking for one statement.
Ex:
How do I get all these scatters in the same one graph, with different colors depending on the value of the variable in each observation?
Gustav KarlssonGustav Karlsson
1 Answer
This is the basic idea of how to do it:
If you store each cluster's y values in their own variable, it makes it easier to plot them all at once.
You can also do it by hand like this:
Dimitriy V. MasterovDimitriy V. Masterov
Not the answer you're looking for? Browse other questions tagged plotgraphstatascatter or ask your own question.
I've created two different clusters, one using kmeans (partitioning method) and one using complete linkage (hierarchical).
The clusters using the hierarchical method:
The clusters using the partitioning method:
This creates two different cluster. Now i want to visually see the clusters created. I want to do a scatter and make it show different color depending of the value of pcm_1 and c1. However, I can only create a scatter checking for one statement.
Ex:
How do I get all these scatters in the same one graph, with different colors depending on the value of the variable in each observation?
![Stata Plots Cluster Aggregates Stata Plots Cluster Aggregates](/uploads/1/2/5/6/125618595/550883217.png)
![Stata Stata](http://wlm.userweb.mwn.de/Stata/grafik/Stata_line-overlay_small.png)
Gustav KarlssonGustav Karlsson
1 Answer
This is the basic idea of how to do it:
If you store each cluster's y values in their own variable, it makes it easier to plot them all at once.
You can also do it by hand like this:
Dimitriy V. MasterovDimitriy V. Masterov