Extreme values of a portfolio of Gaussian processes and a trend

Hüsler, Jürg ; Schmid, Christoph

In: Extremes, 2005, vol. 8, no. 3, p. 171-189

Ajouter à la liste personnelle
    Summary
    We consider the extreme values of a portfolio of independent continuous Gaussian processes $$\sum_{i=1}^{k}w_{i}X_{i}(t)$$ ( $$w_{i} \in \mathbb{R}, \; k \in \mathbb{N}$$ ) which are asymptotically locally stationary, with expectations $$E[X_{i}(t)]=0$$ and variances $$Var[X_{i}(t)]=d_{i}t^{2H_{i}}$$ $$(d_{i} \in \mathbb{R^{+}}, 00$$ with $$\beta > H_{i}$$ . We derive the probability $$P\{ \sup_{t>0} \sum_{i=1}^{k}w_{i}X_{i}(t) -ct^{\beta} > u \}$$ for $$u \rightarrow \infty$$ , which may be interpreted as ruin probability