TABLE 2.

Simulation results for the analysis of $P_{0,2}({\tau }_{0.6})$ and $P_{0,2}^{\prime }({\tau }_{0.6})$, where τ_0.6 is the 60th percentile of the follow‐up time, based on the standard approach that ignores the within‐cluster dependence (naïve) and the proposed method with (i) the influence function–based variance estimator (IF) and (ii) the nonparametric cluster bootstrap (CB)

			$P_{0,2}({\tau }_{0.6})$				$P_{0,2}^{\prime }({\tau }_{0.6})$
n	$F_M$	Method	Biasa	MCSDa	ASEa	CP	Biasa	MCSDa	ASEa	CP
20	$\mathcal{U}[5,15]$	Naïve	0.160	3.657	3.040	0.904	−0.939	3.656	3.033	0.888
		IF	0.160	3.657	3.348	0.921	0.077	3.963	3.651	0.924
		CB	0.160	3.657	3.378	0.916	0.077	3.963	3.663	0.920
	$\mathcal{U}[10,30]$	Naïve	0.116	2.731	2.146	0.869	−0.940	2.740	2.140	0.854
		IF	0.116	2.731	2.679	0.935	0.078	2.978	2.899	0.935
		CB	0.116	2.731	2.693	0.940	0.078	2.978	2.899	0.933
40	$\mathcal{U}[5,15]$	Naïve	0.015	2.360	2.143	0.935	−1.060	2.364	2.140	0.896
		IF	0.015	2.360	2.399	0.955	0.027	2.592	2.635	0.953
		CB	0.015	2.360	2.407	0.957	0.027	2.592	2.636	0.953
	$\mathcal{U}[10,30]$	Naïve	0.035	1.956	1.513	0.866	−1.020	1.943	1.509	0.818
		IF	0.035	1.956	1.915	0.936	−0.011	2.100	2.075	0.937
		CB	0.035	1.956	1.919	0.941	−0.011	2.100	2.075	0.936
80	$\mathcal{U}[5,15]$	Naïve	−0.063	1.745	1.513	0.913	−1.152	1.738	1.510	0.845
		IF	−0.063	1.745	1.714	0.943	−0.084	1.894	1.885	0.949
		CB	−0.063	1.745	1.716	0.945	−0.084	1.894	1.885	0.948
	$\mathcal{U}[10,30]$	Naïve	0.076	1.436	1.073	0.856	−0.972	1.433	1.070	0.775
		IF	0.076	1.436	1.369	0.939	0.045	1.543	1.487	0.942
		CB	0.076	1.436	1.372	0.940	0.045	1.543	1.488	0.945

Abbreviations: ASE, average estimated standard error; CP, coverage probability; n, number of clusters; $F_M$, discrete uniform distribution of the cluster size; MCSD: Monte Carlo standard deviation of the estimates.

Note. Results under right censoring.

^aIndicates × 10².