Random Forests for time-fixed and time-dependent predictors: The DynForest R package

TL;DR

DynForest is an R package that extends random forests to handle complex time-dependent predictors, including endogenous and error-prone data, for various outcome types, with detailed user guidance.

Contribution

The paper introduces DynForest, a novel R package that models time-dependent predictors in random forests, including endogeneity and measurement error, with flexible feature extraction.

Findings

Handles endogenous and error-prone time-dependent predictors.

Provides variable importance and minimal depth measures.

Supports multiple outcome types including continuous, categorical, and survival.

Abstract

The R package DynForest implements random forests for predicting a continuous, a categorical or a (multiple causes) time-to-event outcome based on time-fixed and time-dependent predictors. The main originality of DynForest is that it handles time-dependent predictors that can be endogeneous (i.e., impacted by the outcome process), measured with error and measured at subject-specific times. At each recursive step of the tree building process, the time-dependent predictors are internally summarized into individual features on which the split can be done. This is achieved using flexible linear mixed models (thanks to the R package lcmm) which specification is pre-specified by the user. DynForest returns the mean for continuous outcome, the category with a majority vote for categorical outcome or the cumulative incidence function over time for survival outcome. DynForest also computes variable importance and minimal depth to inform on the most predictive variables or groups of variables. This paper aims to guide the user with step-by-step examples for fitting random forests using DynForest.