A carregar...
Publicação
An application of clustering analysis to the comparison of mortality rates
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.56 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
This work provides the study of dissimilarity between mortality rates in 116 countries.
Mean Shift Algorithm and Principal Component Analysis processed efficiently to
classify countries into clusters, which show the reduction of the rate of mortality in the
last twenty years. The result reveals an evident difference in the children and adult
groups among researched countries, while it is more difficult to classify the rate for the
seniors. Although the mortality rate will gradually reduce through time, there are two
distinguishable patterns for developing countries having high rates: in Africa, where the
value is higher for children, and in East Asia, where the survival probability for the adult
is lower. The main reasons for the high mortality value in researched countries are
natural disasters, global pandemics, and low-quality life. We also provided an
application of the methodology by estimating the expected value and standard deviation
of survival probability in Portugal and Vietnam in the next ten years.
Descrição
Mestrado Bolonha em Actuarial Science
Palavras-chave
Mortality rate Life insurance Clustering Mean shift Principal Component Analysis
Contexto Educativo
Citação
Xuan, Tung Nguyen (2022). “An application of clustering analysis to the comparison of mortality rates”. Dissertação de Mestrado. Universidade de Lisboa. Instituto Superior de Economia e Gestão
Editora
Instituto Superior de Economia e Gestão