Skip to main content

Table 3 Cost coefficients for different types of misclassification

From: Power and sample size calculations in the presence of phenotype errors for case/control genetic association studies

K R* p C θ C φ
0.005 0.5 0.05 0.01 540.29
   0.15 0.01 458.99
  1 0.05 0.01 478.32
   0.15 0.01 432.67
  2 0.05 0.01 440.18
   0.15 0.01 415.60
0.05 0.5 0.05 0.09 51.59
   0.15 0.10 43.82
  1 0.05 0.08 45.67
   0.15 0.10 41.31
  2 0.05 0.08 42.03
   0.15 0.10 39.68
  1. The column heading for this table are as follows: K = prevalence; R* = ratio of controls to cases; p = SNP minor allele frequency in affected population; C θ = Cost coefficient corresponding to misclassification parameter θ – this is a lower bound of the percent increase in sample size necessary to maintain constant asymptotic power for every 1% increase in θ C φ = Cost coefficient corresponding to misclassification parameter φ – this is a lower bound of the percent increase in sample size necessary to maintain constant asymptotic power for every 1% increase in φ. The cost coefficients are computed using equation (1).