Banca de DEFESA: FELIPE HENRIQUE DE SOUZA

Uma banca de DEFESA de DOUTORADO foi cadastrada pelo programa.
STUDENT : FELIPE HENRIQUE DE SOUZA
DATE: 25/11/2025
TIME: 09:00
LOCAL: Videoconferência via Gerência de Redes do CCET/UFRN
TITLE:

Projection with prediction interval for the population of Brazilian states by sex and age groups: 2025 to 2070

KEY WORDS:

demographic components; population projection; prediction interval; uncertainty estimation.

PAGES: 120
BIG AREA: Ciências Sociais Aplicadas
AREA: Demografia
SUBÁREA: Componentes da Dinâmica Demográfica
SUMMARY:

Population projections are essential tools for planning public policies and for anticipating the demand for goods and services. The Brazilian Institute of Geography and Statistics (IBGE) has produced population projections for Brazil since the 1970s. The United Nations (UN) also periodically releases projections for all countries; traditionally, these have been based on deterministic models. Beginning in 2010, however, the UN introduced probabilistic, or interval, projections generated using stochastic models, allowing for the quantification of uncertainty in demographic forecasts. The Bayesian models developed by the Center for Statistics and the Social Sciences at the University of Washington form the methodological foundation of the World Population Prospects (WPP), a biennial revision of global population projections prepared by the UN. The adoption of Bayesian methods enables the modeling of replacement-level changes in demographic components and enhances forecasting for population projections. The aim of this thesis is to apply these same Bayesian models to project the population of each Brazilian state and the Federal District, disaggregated by sex and age group, up to the year 2070. This projection horizon coincides with that adopted in the most recent IBGE revision, published in 2024, which serves as the benchmark for comparison with the results presented here. In addition to generating projections for all states, a validation test was conducted using data up to 2010 for the states of São Paulo and Roraima—the largest and smallest populations in Brazil, respectively. These data were used to project populations for 2020, and the results were compared with observed figures. The validation demonstrated that the method is suitable for application at subnational levels, such as states. The projections indicate that by 2030, the states of Rio Grande do Sul and Rio de Janeiro will experience population decline, while Roraima will be the only state not expected to undergo population reduction within the projected period.

COMMITTEE MEMBERS:
Externo à Instituição - MÁRCIO MITSUO MINAMIGUCHI - IBGE
Externo à Instituição - EVERTON EMANUEL CAMPOS DE LIMA - UNICAMP
Externo ao Programa - 1781198 - FIDEL ERNESTO CASTRO MORALES - nullPresidente - 1346605 - FLAVIO HENRIQUE MIRANDA DE ARAUJO FREIRE
Interno - 2002253 - MARCOS ROBERTO GONZAGA
Interno - 1045286 - VICTOR HUGO DIAS DIOGENES