分野別セミナー

Trimming, i.e. removing the largest summands of a sum of identically distributed (iid) random variables, has a long tradition to prove limit theorems which are not valid if one considers the untrimmed sum - one example is the strong law of large numbers for random variables with an infinite mean or in case of ergodic transformations Birkhoff's ergodic theorem.

In this talk I will first give a background about trimmed strong laws of large numbers for iid random variables and compare these results with some random variables obtained from ergodic transformations, for example piecewise expanding interval maps and subshifts of finite type. If time allows I will also give some insides into the proof for the dynamical systems results using the transfer operator method.