1% Author: Simon-Pierre Boucher — simon-pierre.boucher@uqo.ca2% generated by scripts/20_evaluate.py — do not edit3\begin{tabular}{lrrrrrrrr}4\toprule5Arm & {Cov. 50\%} & {Cov. 80\%} & {Cov. 90\%} & {Cov. 95\%} & {Width 90\% (log)} & {Pinball (log)} & {CRPS (log)} & {n\_with\_interval} \\6\midrule7A (hedonic) & \num{0.489} & \num{0.869} & \num{0.909} & \num{0.939} & \num{1.001} & \num{0.065} & \num{0.133} & \num{96935} \\8A (hedonic spatial) & \num{0.423} & \num{0.834} & \num{0.886} & \num{0.952} & \num{1.012} & \num{0.065} & \num{0.134} & \num{8988} \\9B (hierarchical) & \num{0.499} & \num{0.874} & \num{0.913} & \num{0.939} & \num{0.998} & \num{0.063} & \num{0.129} & \num{96935} \\10C (catboost) & {--} & {--} & {--} & {--} & {--} & {--} & {--} & \num{0} \\11C (gbm) & {--} & {--} & {--} & {--} & {--} & {--} & {--} & \num{0} \\12C (gbm spatial) & {--} & {--} & {--} & {--} & {--} & {--} & {--} & \num{0} \\13C (xgboost) & {--} & {--} & {--} & {--} & {--} & {--} & {--} & \num{0} \\14D (avm) & \num{0.446} & \num{0.868} & \num{0.896} & \num{0.936} & \num{0.830} & \num{0.058} & \num{0.119} & \num{96935} \\15D (avm spatial) & \num{0.425} & \num{0.856} & \num{0.907} & \num{0.938} & \num{0.741} & \num{0.058} & \num{0.120} & \num{8988} \\16\midrule17E (single llm) & \num{0.463} & \num{0.762} & \num{0.887} & \num{0.912} & \num{0.562} & \num{0.055} & \num{0.114} & \num{80} \\18E (single llm small) & \num{0.417} & \num{0.670} & \num{0.777} & \num{0.813} & \num{0.464} & \num{0.064} & \num{0.129} & \num{300} \\19F (independent) & \num{0.512} & \num{0.775} & \num{0.887} & \num{0.900} & \num{0.590} & \num{0.053} & \num{0.109} & \num{80} \\20F (independent small) & \num{0.420} & \num{0.680} & \num{0.787} & \num{0.840} & \num{0.451} & \num{0.061} & \num{0.124} & \num{300} \\21G (debate r2) & \num{0.475} & \num{0.787} & \num{0.887} & \num{0.900} & \num{0.602} & \num{0.053} & \num{0.109} & \num{80} \\22G (debate r2 small) & \num{0.427} & \num{0.703} & \num{0.807} & \num{0.860} & \num{0.489} & \num{0.062} & \num{0.126} & \num{300} \\23H (debate skeptic) & \num{0.525} & \num{0.800} & \num{0.875} & \num{0.912} & \num{0.604} & \num{0.053} & \num{0.109} & \num{80} \\24I (placebo) & \num{0.487} & \num{0.812} & \num{0.887} & \num{0.887} & \num{0.590} & \num{0.052} & \num{0.108} & \num{80} \\25J (cost matched) & \num{0.475} & \num{0.787} & \num{0.863} & \num{0.900} & \num{0.561} & \num{0.054} & \num{0.112} & \num{80} \\26\bottomrule27\end{tabular}28