Smart Growth and Vulnerability to Crisis in the EU Regions - Evaluation with Logistic Regression

Małgorzata Markowska

Abstract

The paper discusses the relations among variables characterizing smart growth of the EU NUTS 2 level regions and sensitivity to economic crisis. First, the procedure identifying the status of being in crises is presented – for individual regions in individ ual years. Initial variables reflect the changes in macroeconomic data. Division by median was used for the kind of standardization. Four groups of regions are identified by means of cluster analysis: sensitive, adaptive, robust and outliers. Then the first selection of significant explaining variables is performed by means of parametric (Student t) and nonparametric (Mann - Whitney U) tests for the two expected values. The final list of important and independent factors is obtained through the s tepwise multivariate logistic regression. The final goal is to find some spheres of smart development having the highest influence on robustness against economic crisis improvement. ROC curves with AUC measures prove models identifying robust regions, and separately vulnerable to crisis regions, to be fairly good.
Author Małgorzata Markowska (EMaT / DRE)
Małgorzata Markowska,,
- Department of Regional Economy
Pages19-30
Publication size in sheets0.55
Book Jedlicka Pavel, Maresova Petra, Soukal Ivan (eds.): Hradec Economic Days Vol. 8(2). Double-blind peer-reviewed proceedings paart II of the international scientific conference Hradec Economic Days 2018, 2018, University of Hradec Kralove, ISBN 9788074357015, 591 p.
Hradec_Economic_Days_Vol. 92).pdf / 44,48 MB / No licence information
Keywords in Englishlogit models, smart growth, vulnerability, classification, NUTS 2 regions
URL https://uni.uhk.cz/hed/site/assets/files/1072/proceedings_2018_2.pdf
Languageen angielski
Score (nominal)15
Score sourceconferenceIndex
Publication indicators WoS Citations = 0
Citation count*
Cite
Share Share

Get link to the record


* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.
Back