M. D. Ugarte Martínez, T. Goicoa, G. Vicente, A. Puranik
Crimes against women (CAW) in India have been continuously increasing lately as reported by the National Crime Records Bureau (NCRB). It has been catalogued as a health problem of major impact by the World Health Organization.
However, there is a lack of spatio-temporal analyses to bring to light a complete picture of the geographical and temporal patterns of CAW. In this talk we focus on the temporal evolution of the geographical pattern of “dowry deaths” in Uttar Pradesh, the most populous state in India during the period 2001-2014. The study of the geographical distribution of crime incidence and its evolution in time pursues identifying specific regions that exhibit high dowry death risks. Different spatial priors are considered analyzing their induced correlation structure and their impact on the final risk estimates. Various priors for the hyperparameters are also revised. Integrated nested Lapplace approximation (INLA) is used for model fitting and inference.
Palabras clave / Keywords: regional data, spatio-temporal analysis, INLA, gender-based violence
Programado
Sesión GT08-2 Inferencia Bayesiana-2: Genomics and Spatial statistics under a Bayesian perspective (BAYES-2). Organizador: Stefano Cabras
29 de mayo de 2018 17:00
Sala 1
