A numerical simulation study was performed to investigate the effects of process noise modeling on state estimates of a low-Earth-orbiting satellite in the global positioning system-based hybrid precise orbit determination (POD) strategy. The use of simulated data enabled the creation of the simulated data with three different correlation types of the unknown forces. The simulation results indicate that the use of a periodic random process noise model in hybrid POD strategy provides better performance than the use of a first-order Gauss-Markov random process. The hybrid POD results obtained when the real tracking data from CHAMP (Challenging Minisatellite Payload) were processed showed improved performance of periodic random process over a first-order Gauss-Markov random process. As a relative measure of orbit accuracy, the hybrid orbit solution was compared against a reference orbit solution, and the results indicate that the radial orbit accuracy of the hybrid solution is within 6 cm, in an rms sense. As an absolute measure of orbit accuracy, satellite laser range residuals were analyzed, and the results indicate that the radial orbit accuracy of the hybrid solution is at the level of 3-cm rms for the CHAMP orbit with 460-km altitude.