The use of minimum rate predictors (MRP) opti- mised at volume-level has been recently proposed for medical image coding, under the name of 3D-MRP algorithm. In this paper, the 3D-MRP technology is investigated for lossless com- pression of natural video sequences, which can be processed as volumetric data. Experimental tests demonstrate the advantage of 3D-MRP for lossless coding when compared to the 3D extension of JPEG2000, the 3D-CALIC, or even the current state-of-the- art HEVC standard. The benefit of the volumetric optimisation techniques is also evidenced by the superior performance of 3D- MRP over other existing MRP algorithms that use the traditional frame-by-frame optimisation approach.