etree: Classification and Regression With Structured and Mixed-Type Data in R

As data continue to grow in complexity, so does the need for models and methods that directly account for their non-trivial structure because any simplification may induce loss of information. Energy Trees have been recently proposed following this principle as a unifying and statistically sound framework for classification and regression with structured and mixed-type covariates. In this work, we provide an illustration of etree, which is the R package where Energy Trees are implemented. We describe its origin, structure, main function, and other important features, such as methods for plotting and making predictions. The package currently covers functional data and graphs as structured covariates. However, thanks to its modular infrastructure, any other type of variable can be easily included.