Abstract: A wind tunnel test was conducted on a rectangular high-rise building. Firstly, based on the third-and the fourth-order statistical moments of wind-pressure time-histories, its non-Gaussian regions were identified. The test results show that on the windward face, the non-Gaussian region was near the edge and the corner areas. For the narrower side face, the non-Gaussian region was near the leading edge. However, the non-Gaussian region was discovered close to the trailing edge where the flow reattachment occurred for the wider side face. Then, by parametric analysis and fluctuating wind-pressure spectrum fitting, a simplified method for peak-factor estimation was derived on the basis of the moment-based Hermite method and the Revised Hermite Model, named Three-parameter Hermite Model. Lastly, the accuracy and efficiency of moment-based Hermite method, Revised Hermite Model, and Three-parameter Hermite Model were tested by the instantaneous extreme value and the histogram of wind-pressure histories. The results show that:compared with the peak factors of the instantaneous extreme values, the peak factors calculated by the moment-based Hermite method were overestimated; while the peak factors of Revised Hermite Model and Three-parameter Hermite Model show a smaller error around 10%. The Probability Density Curves (PDCs) of Three-parameter Hermite Model are in a good agreement with the ones of Revised Hermite Model, which show a better fit to the histogram of wind-pressure histories.