Pseudorandom test of analog and mixed-signal circuits provides a low-cost test solution, however its application has been restricted to linear circuit testing. This paper presents an efficient pseudorandom test method for nonlinear circuits. Our method uses a simplified Volterra series model to characterize nonlinear behaviors of Devices Under Test (DUTs) accurately with low complexity algorithm. A multilevel pseudorandom sequence is used to excite DUTs over a wide range of frequencies and generate a spread-spectrum output response. The cross spectral density of the input test pattern and output response is computed to estimate the parameters of Volterra series and predict the performance of DUTs. In addition, our method can be used to compensate nonlinear errors of DUTs and improve performance. The mathematical background and simulation results are presented to validate the proposed method.