Correlations are of great importance in the preliminary design of a gas turbine engine. They are useful to determine the efficiency and thus the Specific Fuel Consumption (SFC) of the engine, before even defining blade geometries. Also, correlations play a very important role in the understanding of the behavior (performance) of the engine, under different operating conditions. Since correlations are obtained from expensive rig tests and cascades, and since cascades cannot represent all situations in an actual engine, the necessity of finding more realistic and cost-effective representations of the actual flow phenomena is becoming more important. A novel approach would be to use CFD as the experimental test cell to generate such correlations and even to extend their limited regime of applicability. In this paper, using a 3-D finite element viscous, compressible, turbulent code, CFD-based correlations for a high-pressure transonic turbine have been created and validated against Cold Flow Turbine Rig test data. This has been done by studying the effect of changing tip clearance, blade speed, stage pressure ratio and vane stagger angle on the performance characteristics of a turbine stage.