Prof. Alexandra Carpentier
Testing theory is a corner stone of mathematical statistics, with many practical applications. While asymptotic and parametric testing is relatively well understood, many additional issues arise in non-parametric or high dimensional models. This is particularly the case when the two hypotheses are composite, i.e. do not consist of a single point. In this case, only few theoretical results are available, and seemingly simple questions remain open. In this talk, I will introduce the problem of composite-composite testing, and present some own (together with co-authors) results in this field.
