This paper illustrates application of the proper orthogonal decomposition (POD) and the autoregressive (AR) model to simulate large wind pressures due to gusts on a low-rise building. In the POD analysis, the covariance of the ensemble of large wind pressures is employed to calculate the principal modes and coordinates. The POD principal coordinates are modeled using the AR process, and the fitted AR models are employed to generate the principal coordinates. The generated principal coordinates are then used to simulate large wind pressures. The results show that the structure characterizing large wind pressures is well represented by the dominant eigenmodes (up to the first fifteen eigenmodes). Also, wind pressures with large peak values are simulated very well using the dominant eigenmodes along with the principal coordinates generated by the AR models.