A0164
Title: Wildfire prediction using zero-inflated negative binomial mixed models: Application to Spain
Authors: Maria-Dolores Esteban - University-Miguel Hernandez of Elche (Spain)
Maria Bugallo - Miguel Hernandez University of Elche (Spain)
Domingo Morales - University Miguel Hernandez of Elche (Spain) [presenting]
Manuel Marey-Perez - Universidade de Santiago de Compostela (Spain)
Abstract: Zero-inflated negative binomial mixed models are adapted to wildfire data where the number of fires is zero in some months and is overdispersed in others because they allow to describe the patterns that explain both the number of fires and their non-occurrence, as well as provide good prediction tools. In addition to model-based predictions, a parametric bootstrap method is applied for estimating mean squared errors and constructing prediction intervals. This type of data is common in the Mediterranean, where the fires are mainly concentrated around the summer months. The statistical methodology and developed software are applied to model and predict the number of wildfires in Spain between 2002 and 2015 by provinces and months.
