1: \begin{abstract}%

2: The Distributional Random Forest (DRF) is a 

3: recently introduced Random Forest algorithm to 

4: estimate 

5: multivariate conditional distributions. 

6: Due to its general estimation procedure, it can be employed to estimate a wide range of targets such as conditional average treatment effects, conditional quantiles, and conditional correlations. 

7: However, only results about the consistency and convergence rate of the DRF prediction are available so far. 

8: We characterize the asymptotic distribution of DRF and develop a bootstrap approximation of it. This allows  

9: us to derive inferential tools for quantifying standard errors and the construction of confidence regions that have asymptotic coverage guarantees.

10: In simulation studies, we empirically validate the developed theory for inference of low-dimensional targets and for testing 

11: distributional differences between two populations. 

12: \end{abstract}

13: