A stochastic method integrating Monte Carlo simulation (MCS) and multivariate regression analysis is presented. The main purpose of this method is to validate a finite element model taking account of variability in test systems and modeling uncertainties. An improved MCS and an iterative procedure during correlation analysis are explained. Simulation results show that the proposed method provides a satisfactory validation outcome with acceptable calculation cost, and different correlation indices generate diverse variation curves of variables. MSC.Nastran is employed with a purpose to show the methodology’s commonality.