The objective of this thesis is the small area estimation of an economic security indicator. Economic
security is a complex concept that carries a variety of meanings. In the literature there is no a formal
unambiguous definition for economic security and in this work we refer to the definition recently
provided for its opposite, economic insecurity, as the “anxiety produced by the possible exposure to
adverse economic events and by the anticipation of the difficulty to recover from them” (Bossert and
D’Ambrosio, 2013). In the last decade interest for economic insecurity/security has grown constantly,
especially since the financial crisis of 2008, but even more in the last year after the economic
consequences due to the Covid-19 pandemic. In this research, economic security is measures through
a longitudinal indicator that takes into account the income levels of Italian households, from 2014 to
2016. The target areas are groups of Italian provinces, for which the indicator is estimated using
longitudinal data taken from EU-SILC survey. We notice that the sample size is too low to obtain
reliable estimates for our target areas. Therefore we resort to some Small Area Estimation strategies
to improve the reliability of the results. In particular we consider small area models specified at area
level. Besides the basic Fay-Herriot area-level model, we propose to consider some longitudinal
extensions, including time-specific random effects following an autoregressive processes of order 1
(AR1) and a moving average of order 1 (MA1). We found that all the small area models used show a
significant efficiency gain, especially MA1 model.
Marino, Mario (2022) Small Area Estimation of Economic Security, [Dissertation thesis], Alma Mater Studiorum Università di Bologna. Dottorato di ricerca in Scienze statistiche , 34 Ciclo. DOI 10.48676/unibo/amsdottorato/10299.