ABSTRACT
Like neural networks, tree regression is a standard data science tool. The conditional distribution framework is especially relevant in the case of tree regression, since the method creates groupings of data with which one can use standard tools like histograms to estimate these distributions. The chapter provides a careful explanation of the methods used to create such groupings (e.g. how to choose the split values), notes the simplicity and ease of interpretation of the results, and compares the results with the historically more common regression tools.
