Stochastic derivative estimation for max-stable random fields

We consider expected performances based on max-stable random fields and we are interested in their derivatives with respect to the spatial dependence parameters of those fields. Max-stable fields, such as Brown–Resnick Smith very popular extremes. focus two most unbiased stochastic derivative estimation approaches: likelihood ratio method (LRM) infinitesimal perturbation analysis (IPA). LRM requires multivariate density field be explicit, IPA necessitates computation for each simulated value. propose convenient tractable conditions ensuring validity cases field, respectively. Obtaining is intricate owing structure Then risk measures, which constitute one several frameworks where our theoretical results can useful. perform a simulation study shows that both well various configurations, provide real case valuable insurance industry.