Skip to main content
. Author manuscript; available in PMC: 2014 May 21.
Published in final edited form as: Stat Med. 2012 Jul 11;31(28):3760–3772. doi: 10.1002/sim.5447

Table 3.

Percent Rejected from Simulations Under the Alternative Hypothesis using Nominal Two-sided 5 percent level. Values preceded with * are significantly larger than the nominal 5% under the Null hypothesis of Table 2. We used the logrank test (Sun’s [1996] implementation). In Simulation Description: SN=simulation number, Scn=scenario, Ê(M)=average number of assessment times in each simulation, eqA= equal Assessment distributions for both treatments (yes/no). See text for complete descriptions of simulations and tests. Simulations where based on 10,000 replications and all tests were applied to the same 10,000 data sets.

SN Simulation Description Percent Rejected
Scn n0 n1 Ê(M) eqA REI pMC PCLT Score wsrMC wsrHLY wsrPCLT
1a MDA 50 50 11.0 yes 45.0 43.1 45.0 46.7 43.8 45.9 44.9
1b MDA 50 50 11.0 no *10.1 43.2 45.0 46.8 43.5 45.2 45.0
1c MDA 5 5 7.8 no *28.1 25.0 27.1 *48.7 24.9 41.1 27.0
1d MDA 5 50 11.0 no 19.9 23.8 11.6 *43.5 22.3 *41.7 12.2
1e MDA 50 5 10.7 no *22.1 40.4 49.2 *39.1 39.4 *36.3 47.4
2a CA 50 50 156.4 yes 39.9 40.5 42.4 44.5 34.6 42.8 42.3
2b CA 50 50 151.9 no *14.4 37.7 39.3 *42.3 29.9 40.3 39.4
2c CA 5 5 15.6 no *23.2 22.0 27.1 *43.8 13.3 34.5 27.3
2d CA 5 50 86.6 no 19.4 18.7 11.2 *39.3 14.5 *37.1 9.8
2e CA 50 5 79.4 no *20.1 *39.9 48.0 *38.2 32.7 *33.9 44.9
3a DPA 50 50 11.0 yes 36.4 36.1 38.1 39.6 25.6 38.7 38.2
3b DPA 50 50 11.0 no *23.7 35.5 37.1 *41.2 30.5 39.0 37.0
3c DPA 5 5 8.2 no *32.0 24.5 26.8 *47.1 22.9 *38.2 26.8
3d DPA 5 50 11.0 no 27.9 20.6 11.4 *40.1 20.0 *38.5 11.6
3e DPA 50 5 10.7 no *27.8 *39.3 *46.9 *36.1 35.3 *32.9 43.8
4 EA 500 500 3.0 no 0.0 *5.2 0.0 0.1 0.0 0.0 0.0