Spatial homogeneity and heterogeneity of energy poverty: a neglected dimension

Abstract

Since the 1970s, a variety of studies has searched for the sociodemographic, housing and economic determinants of energy poverty. A central question, however, has not been answered by any of the previous studies: what are the national-level determinants, i.e. the determinants that homogeneously provoke a high level of energy poverty in all areas of a country? What are the neighbourhood-specific determinants, i.e. the characteristics that have a heterogeneous impact across the neighbourhoods of a country? This study seeks to answer these questions by analysing the level of energy poverty, the percentage of households' disposable income spent on energy expenditure, in 2473 neighbourhoods of the Netherlands in 2014. By employing a semiparametric geographically weighted regression analysis, the effects of two of the determinants of energy poverty are found to be spatially homogeneous: (i) percentage of low-income households and (ii) percentage of pensioners. The results indicate that the impacts of six of the determinants are spatially heterogeneous: (i) household size, (ii) percentage of unemployment, (iii) building age, (iv) percentage of privately rented dwellings, (v) number of summer days and (vi) number of frost days. Subsequently, the effects of spatially homogeneous and heterogeneous determinants are estimated and mapped; the results are discussed and some policy implications are proposed.

EU Cohesion Policy can't buy me love? Exploring the regional determinants of EU image

“…The latter models, additionally, could provide an insight on location-specific determinants of negative image on EU, which could not be gained otherwise. The geographically weighted models, due to such advantages, have been previously used by a variety of researchers in studies across various disciplines and research topics from criminology (Stein, Conley, & Davis, 2016), poverty (Vaziri, Acheampong, Downs, & Majid, 2018), prenatal care (Shoff, Yang, & Matthews, 2012) to household gas and electricity consumption as well as energy poverty (Mashhoodi et al, 2019a(Mashhoodi et al, , 2019b, and real estate (Geniaux & Napoléone, 2008).…”

Lossy compression of Earth system model data based on a hierarchical tensor with Adaptive-HGFDR (v1.0)

“…The huge data volume brings considerable challenges to the data computation, storage, and analysis on ordinary PCs, which will further limit the research and application of ESMD. Lossy compression, which focuses on saving large amounts of data space by approximating the original data, is considered an alternative solution to meet the challenge of the large data volume (Baker et al, 2016;Nathanael et al, 2013). However, ESMD, as a comprehensive interaction of Earth system variables at different aspects of space, time, and attributes, show significant multidimensional coupling correlations (Runge et al, 2019;Mash-Published by Copernicus Publications on behalf of the European Geosciences Union.…”

“…The mixture of different coupling correlations then leads to complex structures, such as uneven distribution, spatial nonhomogeneity, and temporal nonstationary, which increase the difficulties in accurately approximating data in lossy compression. Thus, developing a lossy compression method that could adequately explore the multidimensional coupling correlations is an important way to reduce the compression error (Moon et al, 2017).…”

The Tensor‐based Feature Analysis of Spatiotemporal Field Data With Heterogeneity

“…The geographical phenomenon is the comprehensive interaction result of various geographical elements. With the different effects of space, time, and attributes, the geographical phenomenon shows the characteristic of heterogeneity, which reflects the uneven distribution, spatially nonhomogeneity and temporally nonstationary of geographical data within an area (Mashhoodi et al, ; Shi et al, ; Wang et al, ). However, the existence of heterogeneity also makes geographic data to be the hybrid mixture of various feature signals, which increases the complexity of analyzing the feature signals from the original geographic data.…”