Hands-On Machine Learning for Algorithmic Trading

Hierarchical clustering

Hierarchical clustering avoids the need to specify a target number of clusters because it assumes that data can successively be merged into increasingly dissimilar clusters. It does not pursue a global objective but decides incrementally how to produce a sequence of nested clusters that range from a single cluster to clusters consisting of the individual data points.

There are two approaches:

1. Agglomerative clustering proceeds bottom-up, sequentially merging two of the remaining groups based on similarity
2. Divisive clustering works top-down and sequentially splits the remaining clusters to produce the most distinct subgroups

Both groups produce N-1 hierarchical levels and facilitate the selection of a clustering at ...