Martinez, et al. [1] analysed monthly numbers of dengue cases as reported in Campinas, southeast Brazil from 1998 to 2008, by SARIMA methods. Assuming X is the original series, they analysed the logarithm of X + 1. The models they proposed and compared are of orders (2,1,2)x(1,1,1)12, (2,1,1)x(1,1,1)12, (1,1,2)x(1,1,1)12, (1,1,1)x(1,1,1)12, (2,1,3)x(1,1,1)12, and (1,1,3)x(1,1,1)12. Using the R software, they chose the SARIMA(2,1,2)x(1,1,1)12 model as the best on the basis of Akaike information criterion, AIC. The result in this work is different: the SARIMA(2,1,1)x(1,1,1)12 model is herein adjudged as the best on the same minimum AIC grounds.
Contribution/ Originality
This paper’s primary contribution is that monthly recorded dengue numbers in Campinas, South east Brazil, follow a SARIMA (2, 1, 1) x (1, 1, 1)12 model. It was previously believed that a SARIMA (2, 1, 2)x(1, 1, 1)12 model was the better model. The Eviews software was used to do the analysis. Residual analysis of the chosen model shows that it is very adequate.
Engineering » International Journal of Natural Sciences Research » Month: 09-2014 Issue: 9