1% Author: Simon-Pierre Boucher — simon-pierre.boucher@uqo.ca2% generated by scripts/20_evaluate.py — do not edit3\begin{tabular}{llrrrrcrrr}4\toprule5Contrast (B vs A) & {Metric} & {$N$} & {A} & {B} & {$\Delta$} & {95\% CI} & {$p$ (boot)} & {$p$ (Holm)} & {$p$ (DM)} \\6\midrule7G (debate r2) vs D (avm) & {Mean $|e|$} & \num{40} & \num{0.197} & \num{0.187} & \num{-0.010} & {[-0.057, 0.043]} & \num{0.700} & \num{0.700} & \num{0.601} \\8G (debate r2) vs F (independent) & {Mean $|e|$} & \num{40} & \num{0.179} & \num{0.187} & \num{0.009} & {[-0.000, 0.021]} & \num{0.097} & \num{0.193} & \num{0.094} \\9G (debate r2) vs E (single llm) & {Mean $|e|$} & \num{40} & \num{0.187} & \num{0.187} & \num{0.001} & {[-0.020, 0.020]} & \num{0.925} & {--} & \num{0.898} \\10G (debate r2) vs E (single llm) & {MdAPE} & \num{40} & \num{0.085} & \num{0.096} & \num{0.011} & {[-0.044, 0.030]} & \num{0.332} & {--} & {--} \\11G (debate r2) vs E (single llm) & {P95 $|e|$} & \num{40} & \num{0.645} & \num{0.662} & \num{0.017} & {[-0.058, 0.087]} & \num{0.631} & {--} & {--} \\12G (debate r2) vs E (single llm) & {$|e|>\log 1.25$} & \num{40} & \num{0.300} & \num{0.300} & \num{0.000} & {[-0.071, 0.071]} & \num{1.000} & {--} & {--} \\13G (debate r2) vs E (single llm) & {Cov. 90\%} & \num{40} & \num{0.750} & \num{0.825} & \num{0.075} & {[0.000, 0.167]} & \num{0.087} & {--} & {--} \\14G (debate r2) vs E (single llm) & {Pinball (log)} & \num{40} & \num{0.065} & \num{0.064} & \num{-0.000} & {[-0.006, 0.005]} & \num{0.897} & {--} & \num{0.860} \\15G (debate r2) vs A (hedonic) & {Mean $|e|$} & \num{40} & \num{0.206} & \num{0.187} & \num{-0.018} & {[-0.054, 0.022]} & \num{0.347} & {--} & \num{0.425} \\16G (debate r2) vs A (hedonic) & {MdAPE} & \num{40} & \num{0.123} & \num{0.096} & \num{-0.027} & {[-0.073, 0.022]} & \num{0.183} & {--} & {--} \\17G (debate r2) vs A (hedonic) & {P95 $|e|$} & \num{40} & \num{0.643} & \num{0.662} & \num{0.019} & {[-0.293, 0.149]} & \num{0.829} & {--} & {--} \\18G (debate r2) vs A (hedonic) & {$|e|>\log 1.25$} & \num{40} & \num{0.325} & \num{0.300} & \num{-0.025} & {[-0.123, 0.086]} & \num{0.723} & {--} & {--} \\19G (debate r2) vs A (hedonic) & {Cov. 90\%} & \num{40} & \num{0.925} & \num{0.825} & \num{-0.100} & {[-0.212, -0.023]} & \num{0.045} & {--} & {--} \\20G (debate r2) vs A (hedonic) & {Pinball (log)} & \num{40} & \num{0.068} & \num{0.064} & \num{-0.004} & {[-0.018, 0.012]} & \num{0.569} & {--} & \num{0.642} \\21G (debate r2) vs C (gbm) & {Mean $|e|$} & \num{40} & \num{0.231} & \num{0.187} & \num{-0.043} & {[-0.090, 0.010]} & \num{0.091} & {--} & \num{0.085} \\22G (debate r2) vs C (gbm) & {MdAPE} & \num{40} & \num{0.151} & \num{0.096} & \num{-0.055} & {[-0.120, 0.034]} & \num{0.086} & {--} & {--} \\23G (debate r2) vs C (gbm) & {P95 $|e|$} & \num{40} & \num{0.808} & \num{0.662} & \num{-0.146} & {[-0.350, 0.077]} & \num{0.126} & {--} & {--} \\24G (debate r2) vs C (gbm) & {$|e|>\log 1.25$} & \num{40} & \num{0.300} & \num{0.300} & \num{0.000} & {[-0.143, 0.167]} & \num{1.000} & {--} & {--} \\25G (debate r2) vs C (gbm) & {Cov. 90\%} & \num{40} & {--} & \num{0.825} & {--} & {{--}} & {--} & {--} & {--} \\26G (debate r2) vs C (gbm) & {Pinball (log)} & \num{40} & {--} & \num{0.064} & {--} & {{--}} & {--} & {--} & {--} \\27F (independent) vs D (avm) & {Mean $|e|$} & \num{40} & \num{0.197} & \num{0.179} & \num{-0.019} & {[-0.062, 0.029]} & \num{0.446} & {--} & \num{0.319} \\28F (independent) vs D (avm) & {MdAPE} & \num{40} & \num{0.096} & \num{0.088} & \num{-0.008} & {[-0.096, 0.045]} & \num{0.619} & {--} & {--} \\29F (independent) vs D (avm) & {P95 $|e|$} & \num{40} & \num{0.698} & \num{0.660} & \num{-0.038} & {[-0.322, 0.084]} & \num{0.260} & {--} & {--} \\30F (independent) vs D (avm) & {$|e|>\log 1.25$} & \num{40} & \num{0.275} & \num{0.275} & \num{0.000} & {[-0.128, 0.147]} & \num{1.000} & {--} & {--} \\31F (independent) vs D (avm) & {Cov. 90\%} & \num{40} & \num{0.900} & \num{0.800} & \num{-0.100} & {[-0.265, 0.024]} & \num{0.231} & {--} & {--} \\32F (independent) vs D (avm) & {Pinball (log)} & \num{40} & \num{0.072} & \num{0.062} & \num{-0.011} & {[-0.024, 0.004]} & \num{0.159} & {--} & \num{0.137} \\33F (independent) vs E (single llm) & {Mean $|e|$} & \num{40} & \num{0.187} & \num{0.179} & \num{-0.008} & {[-0.025, 0.005]} & \num{0.314} & {--} & \num{0.212} \\34F (independent) vs E (single llm) & {MdAPE} & \num{40} & \num{0.085} & \num{0.088} & \num{0.003} & {[-0.053, 0.022]} & \num{0.751} & {--} & {--} \\35F (independent) vs E (single llm) & {P95 $|e|$} & \num{40} & \num{0.645} & \num{0.660} & \num{0.015} & {[-0.069, 0.037]} & \num{0.533} & {--} & {--} \\36F (independent) vs E (single llm) & {$|e|>\log 1.25$} & \num{40} & \num{0.300} & \num{0.275} & \num{-0.025} & {[-0.086, 0.000]} & \num{0.551} & {--} & {--} \\37F (independent) vs E (single llm) & {Cov. 90\%} & \num{40} & \num{0.750} & \num{0.800} & \num{0.050} & {[0.000, 0.128]} & \num{0.219} & {--} & {--} \\38F (independent) vs E (single llm) & {Pinball (log)} & \num{40} & \num{0.065} & \num{0.062} & \num{-0.003} & {[-0.008, 0.001]} & \num{0.116} & {--} & \num{0.040} \\39\bottomrule40\end{tabular}41