Alternative Method for Parametric Estimation of Generalised Extreme Value Distribution

The generalized extreme-value (GEV) distribution has found wide application for describing annual floods, rainfall, wind speeds, wave heights, snow depth, and other extreme events. In this paper the polynomial regression models are introduced for estimating the shape of the GEV distribution for using in the flood frequency analysis. Monte Carlo simulations are conducted to illustrate the performance of using the polynomial regression models compared with the model suggested by Hosking et al. [5]. The results show that the polynomial regression model of order 5 is generally better than the other models and this makes it the best option for estimating the parameters of GEV distribution.