Erica Castilho Rodrigues

Bayesian spatial models with a mixture neighborhood structure

In Bayesian disease mapping, one needs to specify a neighborhood structure to make inference about the underlying geographical relative risks. We propose a model in which the neighborhood structure is part of the parameter space. We retain the Markov property of the typical Bayesian spatial models: given the neighborhood graph, disease rates follow a conditional autoregressive model. However, the neighborhood graph itself is a parameter that also needs to be estimated. We investigate the theoretical properties of our model. In particular, we investigate carefully the prior and posterior covariance matrix induced by this random neighborhood structure, providing interpretation for each element of these matrices.

Permeabilidade estatal e atores privados na oferta da educação infantil: o que dizem os Planos Municipais de Educação (2014-2024)?

In this article we present the results of a research that sought to map the trends concerning the ways of supplying early childhood education, in the childcare stage, described in Municipal Education Plans, either formulated and/or reformulated in 2015 under the determination of Article 8, Law 13005/2014, which establishes the National Education Plan for the period from 2014 to 2024, as well as possible clarifying factors for the offer modality. The municipalities were selected through stratified random sampling. The research made use of techniques of descriptive statistics, content analysis and chi-square test to analyze the gathered data. The elements resulting from the study point out to the dispersion of the offer to actors located outside the state apparatus, allowing the statement that this reasoning tends to be a permanent policy in the country, with possible different effects on the education, reliant on the legal nature of the partner entity, since it is specified in Plans of State, not of government.

Interpreting the correlation structure of spatial models for multivariate lattice data

There is an increasing interest on studying multivariate spatial problems, in which several attributes are measured in the same spatial location. The increasing collection of large geographical databases is stimulating the work with this type of data. The complexity of the phenomenon under study has led to the adoption of a hierarchical model with many layers of latent underlying random effects structured as a spatially oriented neighborhood graph. The large number of latent random parameters naturally leads to a Bayesian approach which models these underlying parameters by imposing stochastic smoothing restrictions. The joint prior distribution is determined by a set of conditional distributions based on intuitively simple geographical neighborhood relationships. However, the implied correlation structure is typically unknown in analytical terms and empirical calculations show that it is complex and counter-intuitive in some cases. In this work, we deal with a flexible model for multivariate measurements...

Multiple neighborhood model for analysis of SIDS space-time data set

Bayesian spatial models for areal data assume a fixed neighborhood structure. The stochastic components are comprised by the observed data at each area and the parameters of their distributions, together with the hyperparameters. In this paper, we allow the spatial structure to be part of the parameter space. We introduce a hierarchically increasing neighborhood structure that retain the Markov property of the usual Bayesian spatial models. We illustrate the use of the model with the analysis of the spatial evolution of Sudden Infant Death rates in North Carolina from 1974 to 2006.