Reproducibility package for When Aggregation Misleads: Bias In Unit-Level Small Area Estimates Of Poverty With Aggregate Data

This paper explores why small area poverty estimates from models at the household level that only use aggregate data as covariates, exhibit systematic bias. The analysis demonstrates that this bias stems from the model’s inability to capture the compete between-household variation in welfare, as they rely solely on covariates aggregated at geographic levels. Through model-based simulations, the paper shows that the bias in these models is minimized when the empirical variability of simulated wefare based on the model is closest to the true empirical variance of welfare at the area level. This finding also has implications for bias in unit-level models.