The purpose of this paper is to propose and investigate a new approach for estimating field measurements based response amplitude operators (RAOs) for offshore structures. The approach is based on applying time-varying autoregressive (TVAR) model. This work is virtually unexplored in offshore engineering field. In the literatures, a number of works have shown that RAO are discrete Fourier transform (DFT). Here, we outline a practical algorithm for TVAR model which uses expectation-maximization (EM) algorithm based Kalman smoother to generate RAOs of surge, heave and pitch motions. The method is then applied to sampled discrete wave as excitation input and motion responses of offshore structures as outputs, generated from simulated field measurements. The proposed approach outlined here has shown tremendous potential in the application of RAO estimation. The results indicate that TVAR model bring enhanced accuracy, less noisy in obtaining RAOs estimates over other Fourier based methods. TVAR model also allows for the creation of time-varying RAOs. Copyright © 2012 by ASME.