THE INTERPRETATION OF FUTURE COSMOLOGICAL DATA SETS E.G. FROM EUCLID WFIRST (WIDE-FIELD INFRARED SURVEY TELESCOPE) AND LSST (LARGE SYNOPTIC SURVEY TELESCOPE) WILL REQUIRE A NEW GENERATION OF ANALYSIS FRAMEWORKS THAT CAN MODEL COSMOLOGICAL OBSERVABLES THEIR SYSTEMATICS AND COVARIANCES WITH THE REQUIRED ACCURACY. MORE PRECISELY THE SUCCESS OF THESE MISSIONS WILL DEPEND ON OUR ABILITY TO MODEL 1) BARYONIC EFFECTS IN THE EVOLUTION OF THE MATTER DENSITY FIELD 2) THE CONNECTION OF GALAXIES AND THEIR DARK MATTER ENVIRONMENT 3) NON-LINEAR STRUCTURE EVOLUTION FOR MASSIVE NEUTRINOS TIME-DEPENDENT DARK ENERGY AND MODIFIED GRAVITY THEORIES 4) OBSERVATIONAL SYSTEMATICS SUCH AS MEASURING SHAPES REDSHIFTS AND POSITION OF GALAXIES AND 5) COVARIANCES/LIKELIHOOD FUNCTIONS FOR THE JOINT ANALYSIS OF MULTIPLE CORRELATED OBSERVABLES THAT ENABLE THE TRANSITION FROM DATA TO MODEL SPACE. WE WILL DEVELOP A FAST HIGH-PRECISION ANALYSIS AND INFERENCE PIPELINE THAT INTERFACES NUMERICAL SIMULATIONS THEORETICAL MODELS INNOVATIONS IN STATISTICAL METHODS AND RESULTS FROM EXISTING DATA TO OVERCOME THESE CHALLENGES AND TO PREPARE FOR THE JOINT ANALYSIS OF FUTURE COSMOLOGICAL MISSIONS. MEMBERS OF OUR TEAM ARE EXPERTS IN NUMERICAL SIMULATIONS THEORY AND DATA ANALYSIS AND WE WILL COMBINE THIS EXPERTISE TO 1) ANALYZE LARGE/HIGH-RESOLUTION NUMERICAL SIMULATIONS TO TARGET UNCERTAINTIES IN MODELING COSMOLOGICAL OBSERVABLES AND THEIR SYSTEMATICS 2) DEVELOP SMART STATISTICAL METHODS THAT COMBINE EXISTING KNOWLEDGE FROM THEORETICAL MODELS WITH NUMERICAL SIMULATIONS TO MINIMIZE THE COMPUTATIONAL COST AND 3) TO TEST THE ACCURACY OF THE DEVELOPED MODELS IN SIMULATED LIKELIHOOD ANALYSES THAT REALISTICALLY ACCOUNT FOR THE COMBINED SYSTEMATICS AND STATISTICS ERROR BUDGET OF WFIRST EUCLID AND LSST.