Skip to main content

Table 1 Estimating p0 and average effect size with different methods using all 100 replicates

From: Estimating the number and size of the main effects in genome-wide case-control association studies

  p0 Estimates Average effect size estimates
  No. times p0 = 1 Mean Std. dev. Mean Std. dev.
Quasi-maximum likelihood 0 0.998425 1.97 × 10-4 0.083 5.10 × 10-3
Maximum likelihood 0 0.998424 1.92 × 10-4 0.083 4.92 × 10-3
Conservative (k = 1) 0 0.998409 1.72 × 10-4 0.083 4.94 × 10-3
Adaptive conservative 0 0.998425 2.12 × 10-4 0.083 5.09 × 10-3
Lowest slope 0 0.998162 2.99 × 10-4 0.082 5.09 × 10-3
Storey-Tibshirani 38 0.984106 1.87 × 10-2 0.052 2.79 × 10-2
Storeya 2 0.972607 2.59 × 10-2 0.051 2.75 × 10-2
Location based estimator 100 1 0 --- ---
  1. aFor Storey's, estimator we used the grid he suggested in his article (0.01, 0.02,...,0.95).
  2. b---, p0 = 1. There is no marker to estimate effect size.