A new alternative for estimating parameters in some widely used mixed linear models is proposed. The approach is called as the layer trainglar transformation estimation. Using the same preliminary transformation for both testing and estimation gives us an integrated set of procedures for the full analysis of mixed linear models. The proposed procedures are computationally stable and space saving.