Interest in transform methods for normalizingtest statistics declined with the advent of computers.More recently, small sample asymptotic methods have beendeveloped to approximate the distributions of complicatedtest statistics. We here propose a generalization to a similarrange of statistics of a classical symmetrizing transform. It isshown to behave in a parametric neighborhood similarly tomethods that exploit exponential tilting. It shows promisefor application to a wide range of inference problems.
展开▼