Análise de sobrevivência aplicada a dados penais de reincidência ao crime da comarca de Primeiro de Maio - PR
The Brazilian prison system is point of great discussion in our current political and social moment. According to the latest reports provided by CPI of Deputados (2009), DEPEN (2014) and IPEA (2015) analysis of repeat offenders individuals is scarce in the literature due to lack of concrete data on...
| Autor principal: | Zanon, Matheus Henrique Pimenta |
|---|---|
| Formato: | Trabalho de Conclusão de Curso (Graduação) |
| Idioma: | Português |
| Publicado em: |
Universidade Tecnológica Federal do Paraná
2020
|
| Assuntos: | |
| Acesso em linha: |
http://repositorio.utfpr.edu.br/jspui/handle/1/7387 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: |
The Brazilian prison system is point of great discussion in our current political and social moment. According to the latest reports provided by CPI of Deputados (2009), DEPEN (2014) and IPEA (2015) analysis of repeat offenders individuals is scarce in the literature due to lack of concrete data on the Brazilian prison system. The numbers show that around 24,5% of individuals are repeat offenders in Brazil, according to the IPEA (2015). Therefore, the objective of this paper is conduct a data analysis in Setor de Carceragem Temporária (SECAT) of the Primeiro de Maio - PR. The data set contains individuals information in the period from december 2009 to december 2015, totaling 356 observations, and the event of interest is the time to recurrence to the first crime for each individual in the observed period. The incorporation of covariates in the study is performed in order to study some characteristics of the observed individuals in relation to time of recurrence crime. Initially, comparisons are made between the Burr XII and Weibull distributions (with and without cure fraction) in the survival analysis to verify what model is best fited to the data in the study. Also comparisons are made between the frequentist inference, by the maximum likelihood estimating, and Bayesian inference. The incorporation of covariates occurs in the distribution Burr XII, which was more flexible com- pared to the other models. To obtain the interval and point estimates for the parameters of the proposed models the SAS software is considered. |
|---|