Identifying groups in data using cluster K-means
Cluster K-means is a nonhierarchical technique to cluster items into groups based on their distances from the group centroid. Minitab uses the MacQueens algorithm to identify groups.
Here, we will look at finding groups of tax revenues for the UK from April 2008 until June 2013 in the data. The value of * for row 49 onwards, next to the dates in the second column, indicates provisional data.
The values are in millions of pounds sterling. We might expect tax revenue patterns to exhibit a measure of seasonality. We will use cluster K-means as a way of grouping the months of the year. As this is expected to be based on the month within a quarter, we will initially set the clusters to three.
In the How ...
Get Minitab Cookbook now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
