Understanding Polynomial Distributed Lag Models: Truncation Lag Implications for a Mosquito-borne Disease Risk Model in Brazil

摘要

Using data for the states of Brazil, we construct a polynomial distributed lag model under dierent truncation lagcriteria to predict reported dengue cases. Accurately predicting dengue cases provides the framework to developforecasting models, which would provide public health professionals time to create targeted interventions forareas at high risk of dengue outbreaks. Others have shown that variables of interest such as temperature andvegetation can be used to predict dengue cases. These models did not detail how truncation lag criteria waschosen for their respective models when polynomial distributed lag was used. We explore current truncation lagselection methods used widely in the literature (marginal and minimized AIC) and determine which of thesemethods works best for our given data set. While minimized AIC truncation lag selection produced the best t toour data, this method used substantially more data to inform its prediction compared to the marginal truncationlag selection method. Finally, the following variables were found to be signicant predictors of dengue in thisregion: normalized dierence vegetation index (NDVI), green-based normalized dierence water index (NDWI),normalized burn ratio (NBR), and temperature. These best predictors were derived from multispectral remotesensing imagery as well as temperature data.