Search

Research on household energy poverty is crucial for addressing basic human needs, highlighting the importance of conducting studies across different countries. This study examines energy poverty in the Visegrad Group countries using 2022 data from the EU Statistics on Income and Living Conditions survey, with the 'inability to keep the home warm' indicator as a measure. The study aims to identify socio-economic factors influencing energy poverty and examine how their impacts differ across the examined countries. Logistic regression, combined with machine learning techniques, was applied for this purpose. Furthermore, the study evaluates the classification results of logistic regression and three machine learning methods—CatBoost, Balanced Random Forests, and Extreme Gradient Boosting—on imbalanced data. It was found that, among the three machine learning methods used, Balanced Random Forests performed the weakest. Logistic regression, effective for our imbalanced data, complements the results and provides deeper insights into the socio-economic factors influencing energy poverty. The study found that Slovakia had the highest percentage of households vulnerable to energy poverty, while Czechia had the lowest. Income, household type, and the presence of disabled individuals were found to be important across all countries. However, other factors varied in their influence from one country to another, highlighting the need for country-specific analyses. Monitoring households' exposure to energy poverty is a challenge for future social policy and the use of different methods provides an in-depth view of this complex issue. [ABSTRACT FROM AUTHOR]

In this paper, we study the differences between forecasts obtained with the classical seasonal ARIMA model and forecasts obtained with the neural network model called iTransformers. The analysis is done on Polish data concerning preterm birth from 2015 to 2020. We compare the results and calculate the effect size to conclude the impact of the obtained differences. [ABSTRACT FROM AUTHOR]

In line with new international financial supervision directives (IFRS9), banks should look at a new set of analytical tools, such as machine learning. The introduction of these methods into banking practice requires reformulation of business goals, both in terms of the accuracy of predictions and the definition of risk factors. The article compares methods for selecting variables and assigning "importance" in statistical and algorithmic models. The calculations were carried out using the example of financial data classification for loan default. The effectiveness of various machine learning algorithms on selected sets of variables was compared. The results of the analyzes indicate the need to revise the concept of the "importance" of a variable so that it does not depend on the structure of the model. [ABSTRACT FROM AUTHOR]

Copyright of Quantitative Methods in Economics / Metody Ilościowe w Badaniach Ekonomicznych is the property of Warsaw University of Life Sciences, Faculty of Applied Informatics & Mathematics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Quantitative Methods in Economics / Metody Ilościowe w Badaniach Ekonomicznych is the property of Warsaw University of Life Sciences, Faculty of Applied Informatics & Mathematics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)