Skip to main content
. Author manuscript; available in PMC: 2018 Sep 1.
Published in final edited form as: Stat Methods Med Res. 2017 Jan 8;27(9):2756–2774. doi: 10.1177/0962280216686131

Table 1.

Simulation results without post-treatment confounders

D’n M N Z NDE
ND^E
NDE RMSE Cov.
rr^
NME
NM^E
NME RMSE Cov.
rr^
NTE
NT^E
NTE RMSE Cov.
rr^
Poi N 100 1 1.641 1.656 0.019 98.3 0.979 0.647 0.662 0.018 94.3 0.690 1.915 1.947 0.035 98.8 0.999
0 1.268 1.286 0.020 98.0 0.978 0.273 0.291 0.019 94.1 0.485
500 1 1.644 1.643 0.005 98.9 1.000 0.652 0.656 0.006 94.8 1.000 1.916 1.919 0.006 98.7 1.000
0 1.264 1.263 0.004 98.6 1.000 0.273 0.276 0.004 94.8 1.000
B 100 1 1.517 1.504 0.018 98.8 0.959 0.464 0.451 0.018 99.8 0.104 1.813 1.792 0.026 99.3 0.933
0 1.349 1.341 0.013 97.9 0.946 0.297 0.289 0.011 99.1 0.104
500 1 1.519 1.512 0.008 99.1 1.000 0.481 0.478 0.006 99.9 0.605 1.823 1.816 0.009 99.7 1.000
0 1.342 1.339 0.006 99.0 1.000 0.304 0.304 0.004 99.6 0.605

NB N 100 1 1.513 1.551 0.043 95.0 0.725 0.699 0.724 0.028 95.6 0.508 1.935 2.002 0.070 96.0 0.900
0 1.236 1.278 0.045 94.5 0.725 0.422 0.451 0.030 94.6 0.508
500 1 1.511 1.533 0.024 96.2 1.000 0.695 0.701 0.008 93.6 0.992 1.932 1.958 0.028 96.6 1.000
0 1.238 1.258 0.021 94.8 1.000 0.421 0.425 0.005 92.4 0.992
B 100 1 1.513 1.554 0.047 96.1 0.547 0.471 0.471 0.013 98.4 0.094 1.804 1.852 0.055 97.6 0.628
0 1.333 1.382 0.054 95.5 0.532 0.290 0.299 0.012 95.9 0.092
500 1 1.524 1.519 0.012 96.8 0.994 0.469 0.458 0.013 98.1 0.555 1.821 1.811 0.015 98.5 1.000
0 1.353 1.354 0.009 96.0 0.996 0.297 0.292 0.006 95.7 0.555

ZIP N 100 1 1.608 1.667 0.063 99.9 0.544 0.675 0.675 0.014 96.2 0.295 1.920 1.953 0.039 99.9 0.896
0 1.244 1.278 0.038 99.1 0.581 0.312 0.285 0.028 99.3 0.036
500 1 1.600 1.591 0.015 99.9 0.996 0.662 0.636 0.027 95.3 0.890 1.908 1.865 0.045 99.6 1.000
0 1.247 1.229 0.020 98.5 0.996 0.308 0.274 0.034 98.3 0.391
B 100 1 1.539 1.565 0.037 99.2 0.501 0.510 0.498 0.016 91.4 0.129 1.848 1.879 0.040 99.9 0.688
0 1.338 1.381 0.048 98.6 0.498 0.309 0.313 0.007 94.0 0.118
500 1 1.507 1.500 0.015 99.3 0.986 0.497 0.488 0.010 88.9 0.989 1.810 1.814 0.014 99.8 1.000
0 1.313 1.326 0.017 98.6 0.993 0.303 0.314 0.012 94.3 0.989

ZINB N 100 1 1.614 1.706 0.099 99.9 0.548 0.525 0.532 0.017 93.1 0.291 1.886 1.958 0.083 99.8 0.786
0 1.361 1.425 0.072 99.2 0.605 0.272 0.252 0.022 96.8 0.099
500 1 1.626 1.626 0.032 99.7 0.984 0.538 0.522 0.020 93.3 0.963 1.888 1.880 0.038 99.6 0.993
0 1.350 1.358 0.027 98.8 0.988 0.263 0.253 0.011 96.6 0.799
B 100 1 1.603 1.614 0.044 99.1 0.421 0.548 0.537 0.021 95.9 0.130 1.942 1.954 0.049 99.8 0.563
0 1.394 1.417 0.043 98.8 0.402 0.340 0.340 0.012 97.2 0.140
500 1 1.535 1.525 0.035 99.3 0.942 0.520 0.524 0.013 96.6 0.883 1.859 1.864 0.040 99.5 0.983
0 1.339 1.340 0.029 98.9 0.930 0.324 0.339 0.017 96.6 0.878

D’n: distribution; M: mediator; N: sample size; NDE: natural direct effect; NME: natural mediation effect; NTE: natural total effect; RMSE: root mean squared error; Cov: 95% CI coverage; rr: empirical rejection rate of the test; Poi: Poisson, NB: negative binomial; ZIP: zero-inflated Poisson; ZINB: zero-inflated negative binomial; N: normal; B: binary