keywords: autocorrelation, Bayesian Models, Box plot, density plot, Likelihood, Posterior distribution, Prior distribution.
Malaria is the leading cause of death of children under 5 years of age in Sub-Saharan Africa and Nigeria in particular, it is an acute febrile illness caused by Plasmodium parasites transmitted through bites of infected female Anopheles mosquitoes. The aim of this study is to estimate the relative prevalence of the disease for this important age-specific group in the months of the year. The Bayesian Simulation Modeling Approach was employed in estimating the rates (unknown parameter π) and relative rates(R) of children infected with malaria for each month of the year. This was done using the Markov Chain Monte Carlo (MCMC) Algorithm implemented on the Windows Bayesian Inference Using Gibbs Sampling (WINBUGS) platform. The Beta and Gamma prior distributions were assumed and the Binomial likelihood as the preferred distribution to estimate the posterior paramaters using simulation. Retrospective aggregated data on registered Malaria patients under 5 years were retrieved from four major hospitals in Makurdi town and used as training dataset for the Algorithm. Results show that the malaria prevalence rate in children under 5 years is (π1 =0.1461) 14.61% in January, a synonymous prevalence rate was observed in the following four months (February, March, April and May), June had a highest prevalence rate with an increase of 23.3% in relative to the previous month. Visually inspecting the density plot and history plot, we have confirmed the model convergence and adequacy. Density plot reflect the target distribution which further validates the prior distribution selected. From the findings, Bayesian modeling via simulation is the most suitable for studying disease spread and adjusting public health measures for disease control especially in the face of incomplete data.