• Home
• About
• Current
• Register
• Archives
• Announcements
• People
• submission
• Login

Spatial data are data observed from different neighboring but non-overlapping areal units. In Nigeria, measles continues to be a leading cause of morbidity and mortality in children. Numerous factors, including spatial (S.P.) and temporal trends impacting the disease's intensity and dissemination, affect the transmission dynamics of measles. Although numerous methods exist for spatially time-to-event data, most S.P. models exclude the interaction effect and spatial autocorrelation. Therefore, this study offers the most effective model for overcoming these constraints and allows researchers to flexibly represent spatial clustering in their disease data. However, this research aims to use a–temporal mode to capture the autocorrelation and interaction effects in measles data. To provide an appropriate model for capturing the autocorrelation and interaction effect in the data across the areal units and time and to determine whether the model investigated above is consistent with capturing the measles data, This work uses simulations and poison distribution to identify the substantial spatiotemporal model and interaction/ autocorrelation influence on area unit data. The findings show that the interaction between the small area unit and period decreases as time increases, while the autocorrelation increases for the five area units. The spatial-temporal model is better fitted to 5 areal units when the period is 50 (5x50); however, in 10 and 15 area units, the interaction between the areal unit and period decreases as time increases, and the autocorrelation also decreases. The spatio-temporal model was better fitted to 10 areal units when the period was 20 (10 × 20) and 15 areal units when the period was 100 (15 × 10). Keywords: Spatial-temporal data, Simulation, Areal units, Modeling, Poison prior