![]() Population data on the number of children under five years of age was combined with spatially explicitly predicted parasitaemia risk to estimate the number of infected children. ![]() A regular grid of 231,865 pixels at 4 km2 spatial resolution covering the whole country was generated to predict the parasitaemia risk at unsampled locations and produce a high-resolution risk map. In particular, the models were fitted on a random sample of 85% of the locations and used the remaining locations to compare model-based predictions with observed prevalence by calculating the Mean Absolute Error (MAE). Model validation was performed on the first two models with the highest probability of having generated the data among those considered. Prediction was carried out using Bayesian kriging based on the model with the best predictive ability.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. Archives
September 2023
Categories |