Tesi etd-11192018-180121
Link copiato negli appunti
Tipo di tesi
Corsi integrativi di II livello
Autore
PERTOSA, SAVERIO PIO
URN
etd-11192018-180121
Titolo
EBLUP Area-level methods for estimating poverty at local level
Struttura
Cl. Sc. Sociali - Scienze Economiche
Corso di studi
SCIENZE ECONOMICHE E MANAGERIALI - Economics (DM 270)
Relatori
Relatore Prof.ssa PORCIANI, LINDA
Parole chiave
- Nessuna parola chiave trovata
Data inizio appello
12/12/2018;
Disponibilità
completa
Riassunto analitico
The paper shows how the joint use of sample surveys - such as the Survey on Income and Living Conditions (EU-SILC) - Census data, Tax declarations, and indices of social capital, can be useful to estimate poverty at the provincial level, within a Fay-Herriot framework of analysis.
I show that transfers have a huge impact in addressing the issue of poverty, but - as stated by Okun - they may create market distortions.
I review survey methodologies for data employed within the models. Four Fay-Herriot analyses are implemented, to show how social transfers affect poverty and how regional poverty lines give rise to non-inflated results. I discuss the robustness of results through an analysis of the distribution of residuals and the area effects by implementing a statistical test.
In the end, both statistical and political conclusions are drawn. The appendix shows why a good survey design is crucial when dealing with Paretian variables, such as income, in our "pre-asymptotic" world.
I show that transfers have a huge impact in addressing the issue of poverty, but - as stated by Okun - they may create market distortions.
I review survey methodologies for data employed within the models. Four Fay-Herriot analyses are implemented, to show how social transfers affect poverty and how regional poverty lines give rise to non-inflated results. I discuss the robustness of results through an analysis of the distribution of residuals and the area effects by implementing a statistical test.
In the end, both statistical and political conclusions are drawn. The appendix shows why a good survey design is crucial when dealing with Paretian variables, such as income, in our "pre-asymptotic" world.
File
| Nome file | Dimensione |
|---|---|
| Tesi_SSSUP.pdf | 1.97 Mb |
Contatta l'autore |
|