An important issue when using Machine Learning algorithms in recent research\nis the lack of interpretability. Although these algorithms provide accurate\npoint predictions for various learning problems, uncertainty estimates\nconnected with point predictions are rather sparse. A contribution to this gap\nfor the Random Forest Regression Learner is presented here. Based on its\nOut-of-Bag procedure, several parametric and non-parametric prediction\nintervals are provided for Random Forest point predictions and theoretical\nguarantees for its correct coverage probability is delivered. In a second part,\na thorough investigation through Monte-Carlo simulation is conducted evaluating\nthe performance of the proposed methods from three aspects: (i) Analyzing the\ncorrect coverage rate of the proposed prediction intervals, (ii) Inspecting\ninterval width and (iii) Verifying the competitiveness of the proposed\nintervals with existing methods. The simulation yields that the proposed\nprediction intervals are robust towards non-normal residual distributions and\nare competitive by providing correct coverage rates and comparably narrow\ninterval lengths, even for comparably small samples.\n