The next section describes the rationale of recursive partitioning methods, starting with single classification and regression trees and moving on to ensembles of trees. Examples are interspersed between the technical explanations and provided in an extra section to highlight potential areas of application. A synthesis of important features and advantages of recursive partitioning methods – as well as important pitfalls – with an emphasis on random forests is given in a later section. For all examples shown here, freely available implementations in the R system for statistical computing (R Development Core Team 2009) were employed. The corresponding code is provided and documented in a supplement as an aid for new users.
