Extremal analysis of short series with outliers: sea-levels and athletics records

The analysis of extreme values is often required from short series which ar e biasedly sampled or contain outliers. Data for sea-levels at two UK east coast sites and data on athletics records for women's 3000 m track races ar e shown to exhibit such characteristics. Univariate extreme value methods p rovide a poor quantification of the extreme values for these data. By using bivariate extreme value methods we analyse jointly these data with related observations, from neighbouring coastal sites and 1500 m races respectivel y. We show that using bivariate methods provides substantial benefits, both in these applications and more generally, with the amount of information g ained being determined by the degree of dependence, the lengths and the amo unt of overlap of the two series, the homogeneity of the marginal character istics of the variables and the presence and type of the outlier.