Data Integration: Techniques and Evaluation

Peter Hackl, Michaela Denk

Publication: Scientific journalJournal articlepeer-review

2 Downloads (Pure)

Abstract

Within the DIECOFIS framework, ec3, the Division of Business
Statistics from the Vienna University of Economics and Business
Administration and ISTAT worked together to find methods to create a
comprehensive database of enterprise data required for taxation microsimulations
via integration of existing disparate enterprise data sources. This
paper provides an overview of the broad spectrum of investigated
methodology (including exact and statistical matching as well as
imputation) and related statistical quality indicators, and emphasises the
relevance of data integration, especially for official statistics, as a means of
using available information more efficiently and improving the quality of a
statistical agency's products. Finally, an outlook on an empirical study
comparing different exact matching procedures in the maintenance of
Statistics Austria's Business Register is presented.
Original languageEnglish
Pages (from-to)135-152
JournalAustrian Journal of Statistics
Volume33
Issue number1&2
Publication statusPublished - 2004

Cite this