UQO Working Paper No. 3 — Hedonic housing price models for the US: parametric, quantile, and machine-learning approaches.
1\quad $R^2$ OLS & 0.6301 & [0.6162, 0.6419] & 0.0067 \\2\quad $R^2$ LightGBM & 0.8088 & [0.7972, 0.8181] & 0.0055 \\3\quad $R^2$ XGBoost & 0.8330 & [0.8224, 0.8415] & 0.0050 \\4\quad $R^2$ XGBoost - $R^2$ OLS & 0.2029 & [0.1934, 0.2130] & 0.0051 \\5\quad $R^2$ XGBoost - $R^2$ LightGBM & 0.0242 & [0.0226, 0.0258] & 0.0008 \\6\quad $R^2$ LightGBM - $R^2$ OLS & 0.1787 & [0.1695, 0.1880] & 0.00507