| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 886.05 KB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Central banks play a crucial role in ensuring the reliability and accessibility of
official statistics, which are fundamental for economic analysis, policymaking, and
public transparency. However, as data volumes expand and statistical dissemination
becomes increasingly digitalized, maintaining consistency across multiple sources
remains a challenge. This study presents the development of an automated tool designed
to validate the consistency of statistical data republished by Banco de Portugal (BdP).
Leveraging APIs from primary sources, including the European Central Bank (ECB)
and Eurostat, the tool systematically compares datasets to detect inconsistencies and
streamline quality control processes. By automating statistical verification, the solution
enhances efficiency, reduces reliance on manual checks, and strengthens data reliability.
The study adopts a Design Science Research (DSR) methodology, integrating
theoretical foundations with practical implementation. The developed tool successfully
identified discrepancies in key statistical series, notably in effective exchange rate
indices, where methodological revisions influenced data values. Despite challenges
related to system integration and adaptability to structural changes in datasets, the tool
demonstrated significant improvements in statistical consistency monitoring.
Future research avenues include incorporating machine learning techniques for
anomaly detection, broadening the tool’s applicability across various economic
indicators, and aligning its framework with evolving international data standards, such
as SDMX. This research underscores the transformative impact of automation in
statistical quality assurance, reinforcing transparency, accuracy, and trust in official
economic data.
Description
Keywords
Data consistency Statistical automation APIs Banco de Portugal Data integration Quality control Official statistics SDMX
Pedagogical Context
Citation
Castor, Ana Beatriz Rodrigues (2025). “Improving data consistency in official statistics: application in the central bank of Portugal”. Dissertação de Mestrado. Universidade de Lisboa. Instituto Superior de Economia e Gestão
Publisher
Instituto Superior de Economia e Gestão