1\begin{table}[p]2\centering\scriptsize3\begin{threeparttable}4\caption{Monte Carlo, exp contam design, $p = 1$, $N = 1,048,576$: scaled gap to $\widehat\theta_N$ (root mean square and median) and 95\% coverage by local sample size $m$.}\label{tab:mc_exp_contam_p1}5\begin{tabular}{@{}lrrrrrrrrr@{}}6\toprule7Estimator & \multicolumn{3}{c}{$m = 16$, $K = 65,536$} & \multicolumn{3}{c}{$m = 64$, $K = 16,384$} & \multicolumn{3}{c}{$m = 256$, $K = 4,096$} \\8 & gap & median & cov & gap & median & cov & gap & median & cov \\9\midrule10(1,0) plain average & 101.734 & 101.718 & 0.000 & 30.226 & 30.203 & 0.000 & 8.279 & 8.269 & 0.000 \\11(1,1) average + 1 Newton & 11.018 & 11.014 & 0.000 & 0.973 & 0.971 & 0.704 & 0.073 & 0.073 & 0.818 \\12(1,2) average + 2 Newton & 0.129 & 0.129 & 0.812 & 0.001 & 0.001 & 0.786 & 5.71e-6 & 5.63e-6 & 0.814 \\13(2,0) half-block jackknife & 12.900 & 12.860 & 0.000 & 4.719 & 4.712 & 0.016 & 0.607 & 0.560 & 0.781 \\14(2,1) half-block + 1 Newton & 0.179 & 0.176 & 0.815 & 0.024 & 0.024 & 0.783 & 4.98e-4 & 3.34e-4 & 0.814 \\15(2,2) half-block + 2 Newton & 3.51e-5 & 3.30e-5 & -- & 6.55e-7 & 5.94e-7 & -- & 5.03e-10 & 1.19e-10 & -- \\16(3,0) quarter-block & 21.466 & 21.450 & -- & 3.068 & 3.042 & -- & 0.361 & 0.253 & -- \\17(3,1) quarter-block + 1 Newton & 0.493 & 0.490 & 0.775 & 0.011 & 0.010 & 0.785 & 2.37e-4 & 6.78e-5 & 0.814 \\18(2,1) valid workers only & -- & -- & -- & -- & -- & -- & -- & -- & -- \\19(3,1) valid workers only & -- & -- & -- & -- & -- & -- & -- & -- & -- \\20analytic Cox--Snell + 1 Newton & 1.432 & 1.429 & 0.602 & 0.242 & 0.241 & 0.778 & 0.022 & 0.022 & 0.816 \\21CSL, 1 round & 1.39e-13 & 0 & -- & 3.13e-13 & 2.27e-13 & -- & 3.19e-13 & 2.27e-13 & -- \\22CSL, 2 rounds & 1.35e-13 & 0 & -- & 3.12e-13 & 2.27e-13 & -- & 3.16e-13 & 2.27e-13 & -- \\23AEE (Lin--Xi) & 120.265 & 120.271 & -- & 33.285 & 33.269 & -- & 8.542 & 8.534 & -- \\24(2,1) stale Hessian & 1.806 & 1.797 & -- & 0.191 & 0.189 & -- & 0.006 & 0.005 & -- \\25centralized MLE & 0 & -- & 0.811 & 0 & -- & 0.786 & 0 & -- & 0.814 \\26\bottomrule27\end{tabular}28\begin{tablenotes}\footnotesize29\item 1000 paired replications per column. gap $=$ root mean square of $\sqrt N\,\lVert\tilde\theta - \widehat\theta_N\rVert$ over all replications (including those with failed local fits); median $=$ its median; cov $=$ empirical coverage of nominal 95\% Wald intervals (observed information at the estimate; `--' when not computed). Monte Carlo standard errors of the gaps and RMSEs are in \texttt{results/processed/summary.csv}; coverage standard error $\approx 0.007$ or less.30\item Fraction of workers with all local fits valid (orders 2 / 3): $m=16$: 1.000 / 1.000; $m=64$: 1.000 / 1.000; $m=256$: 1.000 / 1.000. Rows `valid workers only' drop the workers with a failed sub-block fit from the pilot (weights renormalized) and coincide with the corresponding rows when no fit failed.31\end{tablenotes}32\end{threeparttable}33\end{table}34