In the past decade conditional autoregressive modelling specifications have found considerable application for the analysis of spatial data. Nearly all of this work is done in the univariate case and employs an improper specification. Our contribution here is to move to multivariate conditional autoregressive models and to provide rich, flexible classes which yield proper distributions. Our approach is to introduce spatial autoregression parameters. We first clarify what classes can be developed from the family of Mardia (1988) and contrast with recent work of Kim et al. (2000). We then present a novel parametric linear transformation which provides an extension with attractive interpretation. We propose to employ these models as specifications for second-stage spatial effects in hierarchical models. Two applications are discussed; one for the two-dimensional case modelling spatial patterns of child growth, the other for a four-dimensional situation modelling spatial variation in HLA-B allele frequencies. In each case, full Bayesian inference is carried out using Markov chain Monte Carlo simulation.

Abstract. We propose a stochastic model for the relationship between the entomologic inoculation rate (EIR) for Plasmodium falciparum malaria and the force of infection in endemic areas. The model incorporates effects of increased exposure to mosquito bites as a result of the growth in body surface area with the age of the host, naturally acquired pre-erythrocytic immunity, and the reduction in the proportion of entomologically assessed inoculations leading to infection, as the EIR increases. It is fitted to multiple datasets from field studies of the relationship between malaria infection and the EIR. We propose that this model can account for non-monotonic relationships between the age of the host and the parasite prevalence and incidence of disease. It provides a parsimonious explanation for the faster acquisition of natural immunity in adults than in children exposed to high EIRs. This forms one component of a new stochastic model for the entire transmission cycle of P. falciparum that we have derived to estimate the potential epidemiologic impact of malaria vaccines and other malaria control interventions.