Application of Transformed Prediction Ellipsoids for Outlier Detection in Multivariate Non-Gaussian Data

Abstract

The technique to construct a transformed prediction ellipsoid based on the bijective multivariate normalizing transformation of non-Gaussian random vector are offered. Application of the transformed prediction ellipsoids and a quantile of the Chi-Square distribution for detecting outliers in multivariate non-Gaussian data on the basis of univariate and multivariate normalizing transformations is considered. The example of outliers detection in the three-dimensional non-Gaussian source dataset is given.