Extremal index: estimation and resampling

Abstract The duration of extremes in time leads to a phenomenon known as clustering high values, with strong impact on risk assessment. extremal index is measure developed within Extreme Value Theory that quantifies the degree values. In this work we will consider cycles estimator introduced Ferreira and (Ann Inst Henri Poincare Probab Stat 54(2):587–605, 2018). A reduced bias based Jackknife methodology be presented. bootstrap technique also considered inference allow obtain confidence intervals. performance analyzed simulation. We found our proposal effective reducing it compares favorably some well-known methods. An application methods real data