Extreme value theory and branching processes

After giving an overview on well known results on extreme values for sequences of independent identically distributed random variables, we turn to a class of models that exhibit a logarithmic correlation structure.

As it turns out a (sometimes hidden) branching structure helps to understand the effect of correlations. In this talk we will focus on one such model which is called branching random walk. I will explain how first and (truncated) second moment computations can be used to determine the order of the maximum.