Beyond the Sample: Extreme Quantile and Probability Estimation

Economic problems such as large claims analysis in insurance and value-at-risk in finance, requireassessment of the probability P of extreme realizations Q. This paper provided a semi-parametricmethod for estimation of extreme (P, Q) combinations for data with heavy tails. We solve the longstanding problem of estimating the sample treshold of where the tail of the distribution starts. This isaccomplished by the combination of a control variate type device and a subsample bootstrap technique.The subsample bootstrap attains convergence in probability, whereas the full sample bootstrap wouldonly provide convergence in distribution. This permits a complete and comprehensive treatment ofextreme (P, Q) estimation.