Back to search

Panteion University of Social and Political Sciences

Estimating poverty and unemployment using small area estimation methods

Abstract

dc:description

Fighting poverty and unemployment are two of the goals of the European Commission, which is clearly emphasized in the “Europe 2020” strategy. Τhe surveys carried out for these social characteristics provide very reliable information at national level but due to the design and the sample size they cannot give reliable estimates in smaller geographical areas of the country, such as the prefectures (NUTS 3) and municipalities. The key solution in order to produce reliable estimates in small areas is to combine data from administrative, Census and survey sources using Small Area Estimation methods (SAE). In Greece the lowest geographical level for which estimates of poverty and unemployment are given is the NUTS 2 level (perifereies). The main goal of this thesis is to develop and produce reliable estimates for the poverty and unemployment in Greece at a lower level of spatial aggregation than the one used so far, that is at the level of sub regions-NUTS 3 (Nomoi), using SAE methods. The target parameters to be estimated are headcount ratio, poverty gap index and unemployment rate of the Greek population at two different times, in 2009 (shortly before the start of the Greek financial crisis) and 2013 (during the crisis)). Τo achieve the above objectives the EBLUP estimator based on the Fay and Herriot (F-H) small area model was adopted, combining survey data from the EU-SILC 2009 and 2013 with auxiliary data derived from the 2001 and 2011 national Greek Census, respectively. Specifically, 19 auxiliary variables from the Greek Census of 2001 and 32 auxiliary variables from the Greek Census of 2011 were considered and analyzed. In order to build the optimal small area model a variable selection process was performed for each of the target parameters. Then, different diagnostic tools applied to assess the fit and the performance of the selected small area models as well as to check the reliability of the results. These diagnostics showed that the final selected models provide a good fit to the data and produce reliable estimates. Τhe research results were particularly encouraging as the application of small area estimation approaches achieved an overall significant efficiency gain both for estimating poverty and unemployment instead of direct estimators. Specifically, results showed a significant reduction in coefficient of variation (CV) of the EBLUP F-H estimators over direct estimators for most domains. The reduction tended to be greater for domains with small sample sizes. Also, the results of estimates of both poverty and unemployment showed significant differences in the map of Greece in 2009 and 2013. The present study contributes to the growing demand for estimates of social characteristics in small geographical areas by developing appropriate SAE models and giving estimates for poverty and unemployment in Greece for the first time at the level of sub-regions-NUTS 3 (nomoi). These estimates allow governments to formulate and target policies and thus allocate funds properly to small areas.

Degree

thesis:*
Grantor dc:publisher
Panteion University of Social and Political Sciences
Year dc:date
2021

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Malapani, Eleni
  • Μαλαπάνη, Ελένη

Subjects

dc:subject × 20

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/50025

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Malapani, Eleni; Μαλαπάνη, Ελένη. Estimating poverty and unemployment using small area estimation methods. Panteion University of Social and Political Sciences, 2021. http://hdl.handle.net/10442/hedi/50025