A method for the estimation of nonlinear parameters of a structural dynamic system under dynamic loadings is developed based on system identification approach. The nonlinear dynamic behavior of the structure is represented by the hysteretic model of Baber and Wen. Estimations of the nonlinear parameters are carried out by the extended Kalman filter with the exponentially age-weighting and weighted global iteration techniques. As an example, a simple beam model was selected. From the numerical results, it has been found that the method with the exponentially age-weighting is more effective than the one with the weighted global iteration to estimate the unknown parameters of the nonlinear structural system.