Machine Learning Improves Cross-border Tax Estimates
Multidisciplinary research has established that VAT-results are in practice six times lower than what it should have been. The new estimates rely on machine learning techniques.
A research team of the Universities from Leiden and Amsterdam (with background Data Science, Economy and e-Law) presented their new estimates in de Journal of the Royal Statistical Society, under the title: 'A data-driven supply-side approach for estimating cross-border Internet purchases within the European Union'.
The authors are: Quinten Meertens, Cees Diks (UvA), Jaap van den Herik and Frank Takes.