The ADETON method : an application for the estimation of regional flow matrices





Faculty/Professorship: Regional Labour Economics  
Author(s): Blien, Uwe; Graef, Friedrich
Title of the Journal: Jahrbuch für Regionalwissenschaft : a publication of the German-speaking section of the Regional Science Association International, Gesellschaft für Regionalforschung
ISSN: 1613-9836
Corporate Body: Gesellschaft für Regionalforschung
Publisher Information: Berlin ; Heidelberg : Springer
Year of publication: 2013
Volume: 33
Issue: 2
Pages: 135-150
Language(s): English
DOI: 10.1007/s10037-013-0078-5
Abstract: 
"A problem encountered in many contexts is the estimation of matrices or tables from aggregate, heterogeneous, incomplete and contradictory information. In this paper the Adeton method is presented which can be used for this estimation task under quite general conditions. The Adeton method was originally developed to estimate flow matrices of regional labour markets in the Multi-Accounting System (MAS). However, it is applicable for many purposes, e.g. to estimate contingency tables or input-output and other flow matrices. Adeton is based on a Bayesian inference model: Given a prior probability distribution on the set of possible matrices and information about the actual matrix consisting of a set of linear equality and inequality constraints, the complete matrix with highest posterior probability is calculated. The advantage of the Adeton approach is that it is possible to specify soft constraints which are obeyed only up to a certain degree. It is shown that Adeton is an estimation method of entropy optimization type and in this respect is a generalization of the well known Iterative Proportional Fitting Algorithm (used in log-linear models) or of the equivalent RAS method (used in input-output analysis)."
Type: Article
URI: https://fis.uni-bamberg.de/handle/uniba/39264
Year of publication: 28. July 2015