The role of major mergers in galaxy evolution remains a key open question. Major pair fraction estimates based on galaxy-galaxy pairs often are in tension with each other and with theoretical expectations, especially during 1<z<3. The lack of rigorous calibrations for observational systematics is one of the reasons behind this tension. To help resolve this, we analyze redshift evolution of the major close-pair fraction (<4:1) among massive (>1E10 Msun) galaxies between 0.5<z<3.0 in the SantaCruz Semi-Analytical Model based mock lightcones with CANDELS-like observational effects. We derive statistical pair fraction calibration corrections for common pair selection functions used in the literature. We find that i) ~35% of the photo-z selected pairs have velocities <500 km/s, ii) stellar-mass errors impact major pair selection with only ~85% (65%) of the true major pairs being selected at z~1 (z~3), iii) the incompleteness correction is ~10% (~20%) at z~2 (z~3) for CANDELS-Wide depth, iv) pair fraction variation rises from 1%(z~1) to 1.5% (z~3). We apply these corrections to the empirical CANDELS data and our preliminary results suggest that the calibrated pair fraction evolution remains flat (at ~10%) during 2<z<3. Our methodology and calibrations will enable us to robustly predict close pairs counts in preparation for JWST and Nancy Grace Roman Space Telescopes and future studies to derive robust major merger history of massive galaxies since z~3 using them.