Large p Small n: Inference for the Mean
Abstract
We present a new result that enables inference for the mean vector of a multivariate normal random variable when the number p of its components is far larger than the number n of sample units and the covariance structure is completely unknown. The result turns out to be a useful tool for the inferential analysis (e.i. confidence region and hypothesis testing) of data up to now mostly studied only within an explorative perspective, like functional data. To this purpose, an application to the analysis of brain vascular vessel geometry is developed and shown.
Origin : Files produced by the author(s)
Loading...