The thesis aims to investigate the relationship between a nation’s Democracy Index, as defined by the EIU, and the economic inequality among its citizens, measured as the income share held by the richest 1% (or alternatively the richest 10%) of the population. The sample used is based on a set of countries from around the world, with observations covering the period from 2010 to 2023. The final dataset was independently constructed by integrating data from various sources, while the models defined are linear regressions (both simple and multiple) for panel data, specifically Fixed Effects and Random Effects models. The estimation techniques implemented, in addition to pooled OLS, are therefore Fixed Effects and Random Effects.
La tesi si pone l'obiettivo di indagare la relazione tra l'Indice di Democrazia di una nazione, definito da EIU, e la disuguaglianza economica tra i suoi cittadini, misurata invece come quota di reddito percepita dall'1% (o in alternativa dal 10%) più ricco della popolazione. Il campione utilizzato si basa su un insieme di paesi provenienti da tutto il mondo, le cui osservazioni coprono il periodo di tempo 2010-2023. Il dataset conclusivo è stato costruito in autonomia, integrando dati provieniti da diverse fonti, mentre i modelli definiti sono quelli della regressione lineare (semplice e multipla) per dati di tipo panel, precisamente modelli a Effetti Fissi e a Effetti Casuali. Le tecniche di stima implementate, oltre alla pooled OLS, sono dunque Fixed effects e Random effects.
Democrazia e disuguaglianza economica: un'analisi longitudinale
BINCOLETTO, GIACOMO
2024/2025
Abstract
The thesis aims to investigate the relationship between a nation’s Democracy Index, as defined by the EIU, and the economic inequality among its citizens, measured as the income share held by the richest 1% (or alternatively the richest 10%) of the population. The sample used is based on a set of countries from around the world, with observations covering the period from 2010 to 2023. The final dataset was independently constructed by integrating data from various sources, while the models defined are linear regressions (both simple and multiple) for panel data, specifically Fixed Effects and Random Effects models. The estimation techniques implemented, in addition to pooled OLS, are therefore Fixed Effects and Random Effects.| File | Dimensione | Formato | |
|---|---|---|---|
|
Bincoletto_Giacomo.pdf
accesso aperto
Dimensione
3.47 MB
Formato
Adobe PDF
|
3.47 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/92928