Simulation Experiment
An experimental simulation is conducted to investigate the output under different sampling strategies with a real world data set.
1. Generate covariates for the finest available census dataset using the worldpop covariate modeling process.
2.Sampling census blocks using different sampling criteria.
3. Estimate per pixel Population density using Random Forest algorithm.
4. Redistribute population based on next available census units.
5. Aggregrate per-pixel population to finer level administrative units and compare it to original counts.
6. Replicate process 3-5 for 100 times for each sampling strategy.