I've updated my #Stata cluster utilities
These utilities add to Stata's cluster capabilities, particularly when clustering from a distance matrix, rather than variables. Stata's "cluster stop" commands do not work in the manner you might expect with distance matrices.
The updates are mainly minor changes to keep up with newer versions of Stata. #clusteranalysis #sociology #quantmethods #datascience
Install:
. net from https://teaching.sociology.ul.ie/statacode
. net install clutils
I just replied to a query re optimal size of cluster solutions, a bit outside my area of expertise.
I pointed at Everitt et al as a classic: https://books.google.fr/books/about/Cluster_Analysis.html?id=htZzDGlCnQYC&redir_esc=y
Bouveyron, Celeux, Murphy & Raftery, much more up to date https://math.unice.fr/~cbouveyr/MBCbook/
My own code to estimate Calinski-Harabasz & Duda-Hart for #Stata (since Stata's built-in code doesn't work from distance matrices, only the raw data):
https://ulsites.ul.ie/sociology/sites/default/files/wp2016-01_0.pdf
My main point was, however, "there is no correct answer"
Cluster analysis comprises a range of methods for classifying multivariate data into subgroups. By organising multivariate data into such subgroups, clustering can help reveal the characteristics of any structure or patterns present. These techniques are applicable in a wide range of areas such as medicine, psychology and market research. This fourth edition of the highly successful Cluster Analysis represents a thorough revision of the third edition and covers new and developing areas such as classification likelihood and neural networks for clustering. Real life examples are used throughout to demonstrate the application of the theory, and figures are used extensively to illustrate graphical techniques. The book is comprehensive yet relatively non-mathematical, focusing on the practical aspects of cluster analysis.
Brendan Halpin
Blue Tyson
Videogamemonkey