Accurate population data at a high spatial resolution is required for a number of applications such as planning service delivery and designing effective vaccination programmes. This data may not be available in situations where a full census is not possible, e.g. within insecure regions, or at post-censal time points where population counts are outdated. Innovative Bayesian statistical methods allow population counts to be predicted across space using population data collected for small, well-defined areas and ancillary geospatial covariates. One main challenge associated with this method is accounting for the impact of population data being collected across different seasons and years, and the mismatch in time between population data and geospatial covariates used to predict population in unsampled regions. Here we quantify the impact of these temporal mismatches under scenarios of urban growth and seasonal work migration using simulated data, and provide statistical solutions that improve the accuracy of population estimation.
No extended abstract or paper available
Presented in Session 203. Demographic Estimation for Monitoring and Decision-making in Sparse-Data Settings