Modelos de séries temporais aplicados à violência doméstica
Violence is the intentional use of physical force or power, real or threatened against itself, against another person, or against a group or community that results in or is likely to result in injury, death, psychological harm, deficiency of development or deprivation. Domestic violence is one form...
| Autor principal: | Santos, Cesar Augusto da Rocha Santiago |
|---|---|
| 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/1/7404 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: |
Violence is the intentional use of physical force or power, real or threatened against itself, against another person, or against a group or community that results in or is likely to result in injury, death, psychological harm, deficiency of development or deprivation. Domestic violence is one form of violence, among many others. The goals of this research is to study the number of reported monthly cases of domestic violence, sexual violence and other violence in the regional health directorates of Londrina and Cornélio Procópio-PR. The data will be obtained from the DATASUS TABNET system, directly from the Ministry of Health's open access data, for the years 2009 to 2014. The study design is of the ecological epidemiological with time series component. The models and techniques of time series will be used to evaluate the temporal evolution of the number of cases reported monthly, in order to model the occurrence of the phenomenon. Box and Jenkins models were used, ARIMA models (Auto Integrated Regressive of Moving Average) in order to understand the evolution of the random variable number of reported cases of domestic violence, sexual and other violence ordered in time, all models were implemented in the R software. Using these methods, the results obtained show that the analyzed data had a good fit with the following models ARIMA (0,1,0) and SARIMA (0,1,0) (1, 0.0) 12, since they returned the lowest value of the Akaike criterion, so it can be understood that the data in question are influenced by the differentiation of the series, that is, direct function of the delay operator of the series in the two cases. In SARIMA as there is 1 auto-regressive parameter in the cyclic component, it means that the current month is influenced by the month itself in the previous year. Mathematical and statistical models applied are good decision support tools, since they make it possible to characterize the problem, exhibiting periodic behaviors and can give support to public policy strategies and health surveillance. |
|---|