33. Quantifying the impact of outlier management techniques on digital country rankings
Invited abstract in session TE-4: OR applications, stream contributed papers.
Thursday, 16:45 - 18:15Room: C105
Authors (first author is the speaker)
| 1. | Zoltán Bánhidi
|
| Department of Economics, Budapest University of Technology and Economics | |
| 2. | Imre Dobos
|
| Economics, Budapest University of Technology and Economics |
Abstract
The objective of our study is to create rankings of European Union (EU) member states based on objective weights that provide a comprehensive overview of their digital and economic development. We also aim to examine the impact of outlier management techniques, such as winsorising, on these rankings. To accomplish this, we utilised a macro-level cross-sectional dataset that comprises the principal dimensions of the Digital Economy and Society Index, as published by the European Commission, along with the GDP per capita and AIC indicators from economic statistics. In one version of the dataset, extreme values in the raw GDP per capita data were treated with winsorising, while in another version, they were left untreated. The efficiency indicators were used to rank EU Member States based on the synthesis of the digital and economic dimensions using decision-theoretic methods, two DEA models, and a TOPSIS model. The rankings aim to characterise the digital-economic strengths of EU countries and the digital divide found within the EU, as well as evaluate the impact of outlier management. The rankings are also compared with those of the original DESI scoring model to test the contribution of macroeconomic indicators to the results.
Keywords
- Optimization in industry, business and finance
Status: accepted
Back to the list of papers