In view of the defect of commonly used method for correcting the parameters of high-purity germanium (HPGe) detector, a new correction method based on statistical uncertainty analysis and Monte Carlo simulation is proposed. The relative importance of the parameters of HPGe detector in the simulation results is analyzed, to find important parameters. The Simple Monte Carlo sampling of important parameters is conducted thousands times, from which the sampled values are used as the detector model's parameters. Then the detection efficiencies of the detector for multiple γ sources located at different spatial points are simulated via Monte Carlo N-Particle Transport Code, Version 5(MCNP5). The best parameters of detector can be determined by the minimum deviation between true efficiencies and simulation efficiencies. Take these best parameters as the detector model’s parameters, the results show that the relative errors between the simulation efficiencies and the true efficiencies for the sources 60Co, 137Cs, 241Am at the calibration points are all less than 0.5%, the relative errors between the simulation efficiencies and the true efficiencies at 3 verification points are all less than 0.5%.