Séries temporais: uma análise dos casos de tuberculose na Paraíba.

Abstract

This term paper aims to describe conceptual aspects and systematize methodological indications of time series analysis, as well as its application in order to predict future scenarios related to the evolution of the number of cases of Tuberculosis in Paraíba, by highlighting some factors that can change the distribution or behavior of this disease in the State. Predicting the future time is a first and clear application of time series analysis. In fact, this analysis provides a prediction of the future values of the series. However, uncertainty about future estimates is greater the further one gets from the present, and unexpected intercurrences are more or less common in human activities, which imply additional risk of unanticipated error. It is noteworthy that the Box-Jenkins model was essential to estimate and predict the future data of tuberculosis cases in State of Paraíba. It resulted in a sequence aspect to the graph from the predicted data, in addition to a considerable value of confidence interval.