Estimation of extreme values from sampled time series

The paper focuses on the development of a method for extreme value estimation based on sampled time series. It is limited to the case when the extreme values asymptotically follow the Gumbel distribution. The method is designed to account for statistical dependence between the data points in a rational way. This avoids the problem of declustering of data to ensure independence, which is a common problem for the peaks-over-threshold method. The goal has been to establish an accurate method for prediction of e.g. extreme wind speeds based on recorded data. The method will be demonstrated by application to both synthetic and real data. From a practical point of view, it seems to perform better than the POT and Gumbel methods, and it is applicable to nonstationary time series.