Tune Random Forest of the 'ranger' Package

Tuning random forest with one line. The package is mainly based on the packages 'ranger' and 'mlrMBO'.

Variable Selection Using Random Forests

Three steps variable selection procedure based on random forests. Initially developed to handle high dimensional data (for which number of variables largely exceeds number of observations), the package is very versatile and can treat most dimensions of data, for regression and supervised classification problems. First step is dedicated to eliminate irrelevant variables from the dataset. Second step aims to select all variables related to the response for interpretation purpose. Third step refines the selection by eliminating redundancy in the set of variables selected by the second step, for prediction purpose. Genuer, R. Poggi, J.-M. and Tuleau-Malot, C. (2015) < https://journal.r-project.org/articles/RJ-2015-018/>.

Models Multivariate Cases Using Random Forests

Models and predicts multiple output features in single random forest considering the linear relation among the output features, see details in Rahman et al (2017).

Extensible, Parallelizable Implementation of the Random Forest Algorithm

Scalable implementation of classification and regression forests, as described by Breiman (2001), .

Autoencoding Random Forests

Autoencoding Random Forests ('RFAE') provide a method to autoencode mixed-type tabular data using Random Forests ('RF'), which involves projecting the data to a latent feature space of user-chosen dimensionality (usually a lower dimension), and then decoding the latent representations back into the input space. The encoding stage is useful for feature engineering and data visualisation tasks, akin to how principal component analysis ('PCA') is used, and the decoding stage is useful for compression and denoising tasks. At its core, 'RFAE' is a post-processing pipeline on a trained random forest model. This means that it can accept any trained RF of 'ranger' object type: 'RF', 'URF' or 'ARF'. Because of this, it inherits Random Forests' robust performance and capacity to seamlessly handle mixed-type tabular data. For more details, see Vu et al. (2025) .

Ordered Random Forests

An implementation of the Ordered Forest estimator as developed in Lechner & Okasa (2019) . The Ordered Forest flexibly estimates the conditional probabilities of models with ordered categorical outcomes (so-called ordered choice models). Additionally to common machine learning algorithms the 'orf' package provides functions for estimating marginal effects as well as statistical inference thereof and thus provides similar output as in standard econometric models for ordered choice. The core forest algorithm relies on the fast C++ forest implementation from the 'ranger' package (Wright & Ziegler, 2017) .

Dirichlet Random Forest

Implementation of the Dirichlet Random Forest algorithm for compositional response data. Trees are grown using a Dirichlet log-likelihood splitting criterion, with maximum likelihood ('MLE') and method-of-moments ('MOM') parameter estimation. Provides averaging-based predictions (average of responses within terminal nodes), parameter-based predictions (expected value derived from the estimated Dirichlet parameters within terminal nodes), and distributional predictions represented as a weighted distribution over the training responses. Out-of-bag estimation and impurity- and permutation-based variable importance are also supported. For more details see Masoumifard, van der Westhuizen, and Gardner-Lubbe (2026, ISBN:9781032903910).

Modified Ordered Random Forest

Nonparametric estimator of the ordered choice model using random forests. The estimator modifies a standard random forest splitting criterion to build a collection of forests, each estimating the conditional probability of a single class. The package also implements a nonparametric estimator of the covariates’ marginal effects.

Random Forests for Longitudinal Data

Random forests are a statistical learning method widely used in many areas of scientific research essentially for its ability to learn complex relationships between input and output variables and also its capacity to handle high-dimensional data. However, current random forests approaches are not flexible enough to handle longitudinal data. In this package, we propose a general approach of random forests for high-dimensional longitudinal data. It includes a flexible stochastic model which allows the covariance structure to vary over time. Furthermore, we introduce a new method which takes intra-individual covariance into consideration to build random forests. The method is fully detailled in Capitaine et.al. (2020) Random forests for high-dimensional longitudinal data.

Geographically Weighted Random Forests

Fits geographically weighted random forest models using spatially localized training neighborhoods and 'ranger' as the random forest engine. Supports fixed-distance and adaptive neighborhoods defined by observation rows or unique spatial locations, including repeated observations at the same location. Provides local predictions and permutation-based variable importance for examining spatial variation in predictive relationships. The geographical random forest approach is described by Georganos et al. (2021) , and the 'ranger' engine by Wright and Ziegler (2017) .