Um estudo sobre a evolução de óbitos por câncer de mama no Brasil usando modelos de séries temporais

Abstract

Breast cancer is the most common type of cancer among women and is the leading cause of death worldwide. It is among the five most incident cancers in Brazil.Given this, it is understood that it isimportant to assess the number of deaths in Brazil since knowing the behavior of the disease is essential for public agencies focused on the health and well-being of the population. Thus, this study aims to use time series techniques to analyze the numberof observations regarding the number of deaths from breast cancer (group ICD-10: Malignant neoplasms in the breast) in Brazil, covering the period from January 1996 to December 2018. For this analysis, the variability of the series was verified and the presence of the trend and seasonality components. The Box-Jenkins methodology was used to model the data, and the series under study was well adjusted using models of the SARIMA class. The comparison between the models considered suitable for the series was performed using the AIC and EQMP. The adjusted model was used to make predictions about future observations in this series. According to this forecast, it was possible to observe that for the following months, the series will maintain the pattern it has been maintaining since the beginning of its observation period: a growing increase in the number of deaths from such disease.