Skip to main content

Table 3 Realized βa, mean bias, mean standard error (SE), coverage probability and percentage of the effects that had a p-value < 0.05 in the simulation where haplotypes 001 and 101 had a modeled parameter estimate of 1.10. Each replicate contained 200 individuals.

From: Estimating effects of rare haplotypes on failure time using a penalized Cox proportional hazards regression model

No penalty
Haplotype Frequency Realizeda Mean bias Mean SE Coverage P < 0.05b
001 0.050 1.151 0.067 0.25 0.90 0.98
101 0.003 0.817 -0.868 8132.96 0.97 0.25
010 0.020 0.031 0.079 0.48 0.90 0.10
011 0.005 0.000 -1.363 3514.27 0.92 0.08
100 0.020 0.025 0.026 0.49 0.94 0.06
110 0.002 0.000 -1.082 13327.13 0.95 0.05
111 0.280 0.005 0.012 0.10 0.80 0.20
Ridge penalty
Haplotype Frequency Realizeda Mean bias Mean SE Coverage P < 0.05b
001 0.050 1.151 -0.012 0.24 0.84 0.99
101 0.003 0.817 -0.568 1.77 0.57 0.08
010 0.020 0.031 0.057 1.56 1.00 0.00
011 0.005 0.000 -0.322 2.91 0.98 0.02
100 0.020 0.025 0.022 1.65 1.00 0.00
110 0.002 0.000 -0.338 1.85 0.86 0.14
111 0.280 0.005 0.005 0.11 0.84 0.16
Difference penalty
Haplotype Frequency Realizeda Mean bias Mean SE Coverage P < 0.05b
0 0 1 0.050 1.151 0.003 0.25 0.88 0.98
1 0 1 0.003 0.817 -0.394 1.67 0.87 0.01
0 1 0 0.020 0.031 0.111 1.43 1.00 0.00
0 1 1 0.005 0.000 -0.049 2.48 0.99 0.01
1 0 0 0.020 0.025 0.078 1.53 1.00 0.00
1 1 0 0.002 0.000 -0.139 2.08 1.00 0.00
1 1 1 0.280 0.005 0.019 0.12 0.86 0.14
  1. a) "Realized" denotes to the median estimate from the simulated data, where the haplotypes were known, and could be used in the Cox model directly; b) For haplotypes 001 and 101 these percentages reflect power, whereas the percentages for the other five haplotypes reflect type I error probabilities.