Detecting Price Outliers in European Trade Data with Backward and Forward Methods

We describe empirical work in the domain of clustering and outlier detection, for the analysis of European trade data. It is our first attempt to evaluate benefits and limitations of an approach for regression analysis called forward search (Atkinson and Riani (2000); Atkinson et al. (2004)), within a concrete application scenario and in relation to a comparable backward method developed in the JRC by Arsenis et al. (2005). We introduce the application context with an example, the available datasets, two well specified tasks or statistical patterns and an initial heuristic approach to compare the forward and backward solutions. In the conclusions we interpret the practical results obtained on the specific example. Details of the statistical methods used are left to the reference literature.

PERROTTA Domenico;  TORTI Francesca; 

2007-10-16

EUM Edizioni

JRC37043

https://publications.jrc.ec.europa.eu/repository/handle/JRC37043,