CMStatistics 2018: Start Registration
View Submission - CMStatistics
Title: Hidden Markov random fields for the spatial segmentation of circular data Authors:  Jose Ameijeiras-Alonso - KU Leuven (Belgium) [presenting]
Francesco Lagona - University Roma Tre (Italy)
Monia Ranalli - Roma Tre University (Italy)
Rosa Crujeiras - University of Santiago de Compostela (Spain)
Abstract: The aim is to present a model for providing a spatial segmentation of circular data according to a finite number of latent classes employing a hidden Markov random field. Under this setting, the data are modelled by a finite mixture of parametric densities, whose parameters vary across space according to a latent Markov random field. As such, it can be viewed as an extension of a mixture model to the spatial setting. Motivated by wildfires data in the Iberian Peninsula, a model based on a mixture of Kato-Jones circular densities is suggested. This model takes into account special features of wildfire occurrence data such as multimodality, skewness and kurtosis. The parameters of the model will vary across space according to a latent Potts model, modulated by geo-referenced covariates.