Predicting tax treaty formation using machine learning: Implications for parliamentary practice

Erokhin, D. (2024). Predicting tax treaty formation using machine learning: Implications for parliamentary practice. In: Day of Parliamentary Research 2024, 20 June 2024, Vienna, Austria.

Preview

Text Dmitry Erokhin. Day of Parliamentary Research 2024.pdf - Published Version Available under License Creative Commons Attribution Non-commercial. Download (798kB) | Preview

Abstract

This study delves into the predictive potential of machine learning algorithms for tax treaty formations between countries, addressing a critical gap in international economic relations. Our research question investigates the capability of machine learning to accurately predict future tax treaty engagements based on economic determinants. We utilized a comprehensive dataset comprising variables such as foreign direct investment, trade volumes, GDP, and geographical distance, applying various classification algorithms, with the random forest algorithm demonstrating superior accuracy. Our methodology included training the model on 2018 data and validating it with 2019 data, successfully identifying 59 country pairs likely to engage in tax treaties. The findings indicate that economic factors coupled with machine learning provide a robust framework for predicting tax treaty formations, which traditional econometric methods fail to match in predictive power. This research innovates by integrating advanced machine learning techniques into the domain of international economic policy, significantly enhancing predictive accuracy and decision making efficiency.
The potential relevance of this research to parliamentary practice is profound, particularly in understanding how new technologies like machine learning can enhance the capacities of parliaments. By equipping policymakers with predictive insights about tax treaty formations, this study aids in better resource allocation and strategic planning in international relations and economic policies. Furthermore, it prompts legislative bodies to consider regulatory frameworks that incorporate technological advancements to improve governance and policy effectiveness in global economic interactions. This research thus not only contributes to academic literature but also serves as a vital tool for legislative and economic strategists, enhancing the proactive capabilities of parliaments in a digitally evolving landscape.

Item Type:	Conference or Workshop Item (Paper)
Research Programs:	Advancing Systems Analysis (ASA) Advancing Systems Analysis (ASA) > Cooperation and Transformative Governance (CAT)
Depositing User:	Luke Kirwan
Date Deposited:	21 Jun 2024 07:46
Last Modified:	21 Jun 2024 07:46
URI:	https://pure.iiasa.ac.at/19823

Actions (login required)

View Item