Assessing the Effect of Climate Variables on the Incidence of Dengue Cases in the Metropolitan Region of Panama City.

The present analysis uses the data of confirmed incidence of dengue cases in the metropolitan region of Panama from 1999 to 2017 and climatic variables (air temperature, precipitation, and relative humidity) during the same period to determine if there exists a correlation between these variables. In addition, we compare the predictive performance of two regression models (SARIMA, SARIMAX) and a recurrent neural network model (RNN-LSTM) on the dengue incidence series. For this data from 1999-2014 was used for training and the three subsequent years of incidence 2015-2017 were used for prediction. The results show a correlation coefficient between the climatic variables and the incidence of dengue were low but statistical significant. The RMSE and MAPE obtained for the SARIMAX and RNN-LSTM models were 25.76, 108.44 and 26.16, 59.68, which suggest that any of these models can be used to predict new outbreaks. Although, it can be said that there is a limited role of climatic variables in the outputs the models. The value of this work is that it helps understand the behaviour of cases in a tropical setting as is the Metropolitan Region of Panama City, and provides the basis needed for a much needed early alert system for the region.

Previsão da concentração de material particulado inalável, na Região da Grande Vitória, ES, Brasil, utilizando o modelo SARIMAX / Inhalable particulate matter concentration forecast, in the Greater Vitória Region, ES, Brazil, using the SARIMAX model

RESUMO Este trabalho objetivou modelar e prever a concentração média diária de material particulado inalável (MP10), na Região da Grande Vitória (RGV), Espírito Santo, Brasil, utilizando o modelo SARIMAX para o período de 01/01/2012 a 30/04/2015. Os dados deste estudo foram do tipo séries temporais de concentrações de MP10 e de variáveis meteorológicas (velocidade do vento, umidade relativa, precipitação pluvial e temperatura), obtidas junto ao Instituto Estadual de Meio Ambiente e Recursos Hídricos (IEMA), sendo escolhida a estação da Enseada do Suá para fazer o estudo de predição e previsão. Baseando-se em indicadores de desempenho de modelagem, verificou-se que o modelo SARIMAX (1,0,2) (0,1,1)7 é o mais acurado entre os estudados, objetivando fazer predições e previsões da qualidade do ar na RGV. Em comparação com os modelos ARMA, o desempenho estatístico do modelo SARIMAX foi superior, no que diz respeito à predição de eventos de qualidade do ar regular. Dentre as variáveis meteorológicas avaliadas, a velocidade do vento e a precipitação pluvial foram significativas e melhoraram o ajuste do modelo. Em termos de previsão da qualidade do ar, os modelos de séries temporais mostraram resultados satisfatórios.

ABSTRACT This study aimed to model and forecast the average daily concentration of inhalable particulate matter (PM10), in the Greater Vitoria Region (GVR), Espírito Santo, Brazil, using the SARIMAX model, for the period from January 1st, 2012 to April 30th, 2015. Data set from the State Environmental Institute was used. The Enseada do Suá station was chosen for purposes of prediction and forecasting. Some meteorological parameters (wind speed, relative humidity, rainfall and temperature) measured at the GVR were taken as explanatory variables of PM10 concentrations. Based on modelling performance indicators, it was verified the SARIMAX model (1.0.2) (0.1.1)7 is the most accurate between the ones studied, purposing to predict and forecast the air quality in the GVR. The statistical performance of the SARIMAX model was better than the ARMA model, with regard to prediction of regular air quality events. Among the evaluated meteorological variables, wind speed and rainfall were significant and improved the model estimated. Regarding to air quality forecasting, the time series models showed satisfactory results.