<abstract xmlns="http://www.w3.org/1999/xhtml"><p>The purpose of this study is to analyze the health status in Romania at regional NUTS3 level together with its influential socio-economic factors. Apart from statistical and classical econometrics which are being used in most studies, a spatial analysis has been conducted in order to determine possible similarities and dissimilarities among regions, accounting for the fact that events taking place in a specific area are interrelated with the events in the neighboring regions. The negative distribution of the dependent variable, life expectancy, involves the use of Quantile Spatial Autoregressive Model which also allows to observe the socio-economic and environmental factor influences in different parts of health status proxy distribution. The analysis has led to the conclusion that greater the gaps between rich and poor, or greater the difference between less versus better educated, the greater the differences in health status and life expectancy are. Hence a need for policies designed to reduce territorial health disparities has been identified across Romania’s counties. Moreover, Computer Vision and Deep Learning techniques have been used in order to showcase data collection for urban green spaces variables given that more than half of the globe population is living in urban areas and urban greenery has a high positive influence on health. Using Deep Learning on this particular matter together with the Quantile Spatial Autoregressive Model is an innovative approach that has the main aim of improving the classical econometric modelling.</p></abstract>

The purpose of this study is to analyze the health status in Romania at regional NUTS3 level together with its influential socio-economic factors. Apart from statistical and classical econometrics which are being used in most studies, a spatial analysis has been conducted in order to determine possible similarities and dissimilarities among regions, accounting for the fact that events taking place in a specific area are interrelated with the events in the neighboring regions. The negative distribution of the dependent variable, life expectancy, involves the use of Quantile Spatial Autoregressive Model which also allows to observe the socio-economic and environmental factor influences in different parts of health status proxy distribution. The analysis has led to the conclusion that greater the gaps between rich and poor, or greater the difference between less versus better educated, the greater the differences in health status and life expectancy are. Hence a need for policies designed to reduce territorial health disparities has been identified across Romania’s counties. Moreover, Computer Vision and Deep Learning techniques have been used in order to showcase data collection for urban green spaces variables given that more than half of the globe population is living in urban areas and urban greenery has a high positive influence on health. Using Deep Learning on this particular matter together with the Quantile Spatial Autoregressive Model is an innovative approach that has the main aim of improving the classical econometric modelling.

A Spatial Econometric Analysis of Health in Romania Augmented with Computer Vision

Proceedings of the International Conference on Applied Statistics

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

The purpose of this study is to analyze the health status in Romania at regional NUTS3 level together with its influential socio-economic factors. Apart from...

A Spatial Econometric Analysis of Health in Romania Augmented with Computer Vision

Publicado en línea: 13 feb 2020

Páginas: 288 - 303

DOI: https://doi.org/10.2478/icas-2019-0026

Palabras clave
regional health disparities, spatial econometrics, computer vision, deep learning, urban greenery

© 2019 Florin-Ionuț Jurchiş, published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

A Spatial Econometric Analysis of Health in Romania Augmented with Computer Vision

Publicado en línea: 13 feb 2020

Páginas: 288 - 303

DOI: https://doi.org/10.2478/icas-2019-0026

Palabras claveregional health disparities, spatial econometrics, computer vision, deep learning, urban greenery

© 2019 Florin-Ionuț Jurchiş, published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Palabras clave
regional health disparities, spatial econometrics, computer vision, deep learning, urban greenery