{
 "record_id": "b922ab81201881b47779769df9e76de174e13df7",
 "eligible": true,
 "short_ref": "Oliveira, Reyes-Bueno, Gonzalez & da Silva (2024/2025)",
 "country": "Brazil",
 "city": "Fortaleza",
 "property_type": "apartments",
 "price_type": "listed prices and actual sales (log price)",
 "n_obs": 43585,
 "sample_period": "2017 and 2021",
 "data_source": "Real Estate Market Observatory, Local Finance Secretary of Fortaleza",
 "methods_compared": [
  "MRA (OLS)",
  "spatial error regression",
  "MARS",
  "M5P",
  "random forest",
  "quantile random forest",
  "GBM",
  "LightGBM",
  "CatBoost",
  "XGBoost",
  "deep learning"
 ],
 "best_method": "XGBoost",
 "metric": "MAPE",
 "best_value": 8.96,
 "hedonic_value": 13.9,
 "test_set": "holdout",
 "features_ai": "none",
 "incremental_ai_gain": null,
 "incremental_ai_metric": "",
 "marginal_effects": "coefficients",
 "notes": "Appendix Table B, stratified random split 32,688 train / 10,897 test: MAPE 8.96% XGBoost, 9.05% RF, 9.25% GBM, 13.90% MRA, 11.02% SER; R2 0.96 for RF/GBM/LightGBM/XGBoost vs 0.88 MRA and 0.94 SER; RMSE in BRL (XGBoost 116,633.77; GBM lowest 113,640.60; MRA 197,400.28); COD 8.98% for RF and XGBoost. XGBoost is the only model with no residual spatial autocorrelation (Moran's I). Coefficients reported for MRA/SER only; ML models are not explained (authors defer interpretation to another study).",
 "confidence": "high"
}