Knowledge of in situ stress magnitude and orientation plays an important role in geological/geotechnical engineering and in the development of energy resources, such as caprock integrity, waste fluid disposal, geological storage of CO2, and geothermal energy extraction. The uncertainty of estimated parameters, especially horizontal stress, from in situ tests such as pressuremeter tests, is a long-existing challenge due to the existence of uncertainties from geomaterial spatial variability, measurement errors, limited information, and modelling method. Therefore, non-unique solutions are often encountered in the pressuremeter interpretation. In this research, a statistical inverse analysis method is proposed to solve this issue by combining the closed-form solution, finite-difference model, and selected optimization algorithms. The objective of the statistical inverse analysis is to determine the optimal parameters while providing the confidence intervals of derived parameters. Random variables generated in the optimization process reproduce the potential parameter uncertainties. The Jacobian matrix and the confidence intervals are derived from the optimization process to evaluate the variability of predicted horizontal stress and ground properties. A workflow that demonstrates a statistical inverse method for analyzing pressuremeter results is presented that helps to quantify the uncertainties of the ground properties and in-situ stress magnitudes and orientations.