Μελέτη των παραγόντων κινδύνου που σχετίζονται με την πανδημία COVID-19 με χρήση τεχνικών πολυμεταβλητής ανάλυσης
Study of risk factors related to the COVID-19 epidemic using multivariate analysis techniques

Master Thesis
Author
Champesi, Ioanna P.
Χαμπεσή, Ιωάννα Π.
Date
2024-10Advisor
Koutras, MarkosΚούτρας, Μάρκος
View/ Open
Keywords
COVID-19 ; Multivariate analysis ; Principal component analysis ; Partial least squares ; Cluster analysis ; Madrid ; Latent variables ; Principal components ; Risk factors ; PandemicAbstract
The COVID-19 pandemic has introduced significant global challenges, requiring detailed analysis of vast datasets to guide public health decisions. This thesis studies the implementation of multivariate analysis methods to explore the non-clinical factors influencing the pandemic's impact in the community of Madrid. While much of the existing research has focused on clinical outcomes and medical interventions, the present study aims at broadening the understanding by examining demographic, socioeconomic, and climatological variables. The research employs Principal Component Analysis, Partial Least Squares Regression, and k-means clustering to process and interpret high-dimensional data. These techniques are suitable for uncovering complex patterns and relationships that may be present in the datasets, thereby shedding light on some of the key factors associated with COVID-19 transmission and severity. Building on the study by Pérez-Segura et al. [66], the thesis demonstrates how these techniques can be applied to real-world data to uncover critical risk factors. The findings emphasise the need for tailored public health strategies, considering the diverse conditions across different regions. Through this exploration, it aims to enhance the statistical power of data analysis, ensuring more accurate and comprehensive insights into the pandemic's multidimensional impact.