| dc.description.abstract | This thesis is a quantitative study based on the data gathered from Eurostat. The thesis
investigates energy poverty by observing several sides of the problem: geographical distribution
in the European Union, cross-country pattern similarities in the EU, and vulnerability of
European households to energy poverty, especially when energy prices are unprecedentedly
high.
The analysis is performed with the help of such statistical methods as Principal Component
Analysis (PCA) and Hierarchical Clustering (HC). According to PCA, the first four Principal
Components out of fourteen are sufficient for the analysis since they explain 79% of the
variance in the data. Later, HC is applied to those four identified Principal Components,
showing that it is optimal to divide the EU countries into seven categories by their
predisposition and susceptibility to risks associated with energy poverty. Further, the translog
regression approach, along with the HC, is adopted to make a model with an interaction term
comprised of the cluster and household electricity price variables to assess the electricity price
elasticity of household energy consumption.
This thesis is inspired by similar studies conducted by Recalde et al. (2019) and Chai et al.
(2021). However, the paper proposes a different way of tracking energy poverty across Europe,
based on social, economic, environmental and energy indicators. The findings of this thesis
suggest that the neighboring counties' sensitivity to energy poverty tends to be similar, and
southern European states are noticeably more vulnerable to the severe effects of energy poverty. | en_US |
