Table 3 Performance of regression tree models using various features to predict the DHS Wealth Index for Philippines and India. The table reports cross-validated \(R^{2}\) and RMSE values
From: Mapping socioeconomic indicators using social media advertising data
Model features | |||||
|---|---|---|---|---|---|
Interpolated DHS Wealth Index | X | X | |||
Facebook features | X | X | X | ||
Log population density | X | X | |||
Regional indicators | X | X | |||
Philippines (N = 1205) | \(R^{2}\) | 0.480 | 0.608 | 0.627 | 0.630 |
RMSE | 50,983 | 44,218 | 43,099 | 42,965 | |
India (N = 28,043) | \(R^{2}\) | 0.652 | 0.563 | 0.691 | 0.728 |
RMSE | 46,810 | 52,502 | 44,149 | 41,394 |