Skip to main content
. 2017 Nov 9;20(1):1–16. doi: 10.1093/biostatistics/kxx058

Table 1.

EPS performance and comparison with MICE on the simulation study (true Inline graphic, 100 simulated data sets)

  Adjustment/imputation Posterior mean for Inline graphic Bias RMSE CI95 width CI95 coverage
Scenario 1: Inline graphic are available in all areas
  Direct adj 0.50 0.00 0.02 0.054 0.90
  EPS adj 0.50 0.00 0.02 0.062 0.95
Naïve case: Ignoring Inline graphic
  NA 0.78 0.28 0.28 0.049 0.00
Scenario 2: Inline graphic are only available in some areas
  Case 2.1: Analysis on Inline graphic EPS adj 0.51 0.01 0.03 0.087 0.89
  Case 2.2 : Analysis on MICE 0.54 0.04 0.05 0.107 0.65
  Inline graphic EPS imput 0.50 0.00 0.03 0.084 0.90
Scenario 3: Inline graphic are NOT directly available, but Inline graphic are available in some areas
extbfSample size Inline graphic
  Case 3.1.1: Analysis on Inline graphic EPS adj 0.59 0.08 0.09 0.094 0.16
  Case 3.2.1: Analysis on MICE 0.62 0.12 0.14 0.132 0.20
  Inline graphic EPS imput 0.57 0.07 0.08 0.087 0.32
Sample size Inline graphic
  Case 3.1.2: Analysis on Inline graphic EPS adj 0.59 0.09 0.10 0.094 0.47
  Case 3.2.2: Analysis on MICE 0.62 0.12 0.14 0.129 0.51
  Inline graphic EPS imput 0.57 0.07 0.08 0.086 0.60
Sample size Inline graphic
  Case 3.1.3: Analysis on Inline graphic EPS adj 0.59 0.09 0.10 0.094 0.52
  Case 3.2.3: Analysis on MICE 0.60 0.10 0.11 0.128 0.53
  Inline graphic EPS imput 0.57 0.07 0.08 0.085 0.65
Sample size Inline graphic
  Case 3.1.4: Analysis on Inline graphic EPS adj 0.52 0.02 0.04 0.088 0.78
  Case 3.2.4: Analysis on MICE 0.56 0.06 0.06 0.119 0.56
  Inline graphic EPS imput 0.51 0.01 0.03 0.083 0.85