In this thesis it is suggested a decomposition by subpopulation of the Bonferoni index. This purpose is attained at first suggesting a definition of the Bonferroni index for data expressed in terms of frequency distribution and then, using the approach proposed by Zenga (Zenga, 2012) for the decomposition of his index, it follow a decomposition by subpopulation of the Bonferroni index in two steps. The Bonferroni and Zenga indexes, as well as the Gini index (Gini, 1914), are implemented in two steps: in the first step, point measures are made, whereas in the second steps, a synthetic index is obtained averaging the point measures. Until now the proposals suggested for the decompositions of the Gini and Bonferroni indexes were based on the decomposition of the synthetic indexes; in this works, the decomposition by subpopulation is based instead on the decompositions of the point measures. Following this approach, in the first part of the suggested procedure, an additive decomposition is obtained and the contributions of each subpopulation to the point inequality measure, evaluated on the whole population, is computed. From this decomposition, other two decomposition terms are deduced: a within term, that is a measure of inequality evaluated within a subpopulation, and a between term that informs on the inequality derived from the comparison between different subpopulation. In the second part of this work, exploiting these decomposition, an additive decomposition is obtained for the synthetic index. Finally, the methodology proposed is applied on data on the net disposable households income provided by Banca d’Italia in 2014 and comparison among inequality indexes are made.

(2016). Scomposizione per sottopopolazioni dell'indice di Bonferroni. (Tesi di dottorato, Università degli Studi di Milano-Bicocca, 2016).

Scomposizione per sottopopolazioni dell'indice di Bonferroni

VALLI, IGOR
2016

Abstract

In this thesis it is suggested a decomposition by subpopulation of the Bonferoni index. This purpose is attained at first suggesting a definition of the Bonferroni index for data expressed in terms of frequency distribution and then, using the approach proposed by Zenga (Zenga, 2012) for the decomposition of his index, it follow a decomposition by subpopulation of the Bonferroni index in two steps. The Bonferroni and Zenga indexes, as well as the Gini index (Gini, 1914), are implemented in two steps: in the first step, point measures are made, whereas in the second steps, a synthetic index is obtained averaging the point measures. Until now the proposals suggested for the decompositions of the Gini and Bonferroni indexes were based on the decomposition of the synthetic indexes; in this works, the decomposition by subpopulation is based instead on the decompositions of the point measures. Following this approach, in the first part of the suggested procedure, an additive decomposition is obtained and the contributions of each subpopulation to the point inequality measure, evaluated on the whole population, is computed. From this decomposition, other two decomposition terms are deduced: a within term, that is a measure of inequality evaluated within a subpopulation, and a between term that informs on the inequality derived from the comparison between different subpopulation. In the second part of this work, exploiting these decomposition, an additive decomposition is obtained for the synthetic index. Finally, the methodology proposed is applied on data on the net disposable households income provided by Banca d’Italia in 2014 and comparison among inequality indexes are made.
ZENGA, MICHELE
Bonferroni index, decomposition by subpopulations
SECS-S/01 - STATISTICA
Italian
23-giu-2016
STATISTICA - 11R
28
2014/2015
open
(2016). Scomposizione per sottopopolazioni dell'indice di Bonferroni. (Tesi di dottorato, Università degli Studi di Milano-Bicocca, 2016).
File in questo prodotto:
File Dimensione Formato  
phd_unimib_068661.pdf

accesso aperto

Descrizione: Tesi dottorato
Tipologia di allegato: Doctoral thesis
Dimensione 742.59 kB
Formato Adobe PDF
742.59 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/114249
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
Social impact