Pure interaction effects unseen by Random Forests
This addresses a specific limitation in Random Forests for statistical modeling, but it is incremental as it modifies existing methods rather than introducing a new paradigm.
The paper tackles the problem that Random Forests perform poorly in capturing certain pure interactions due to limitations of the CART criterion, and shows that alternative partitioning schemes improve model fitting in such scenarios, as validated in simulations and real datasets.
Random Forests are widely claimed to capture interactions well. However, some simple examples suggest that they perform poorly in the presence of certain pure interactions that the conventional CART criterion struggles to capture during tree construction. Motivated from this, it is argued that simple alternative partitioning schemes used in the tree growing procedure can enhance identification of these interactions. In a simulation study these variants are compared to conventional Random Forests and Extremely Randomized Trees. The results validate that the modifications considered enhance the model's fitting ability in scenarios where pure interactions play a crucial role. Finally, the methods are applied to real datasets.