Skip to main content
. Author manuscript; available in PMC: 2020 May 22.
Published in final edited form as: Stat Methods Med Res. 2017 May 8;26(4):1630–1640. doi: 10.1177/0962280217708664

Table 1.

Data generating distributions for simulation.

Simulation logit E[Y|A, W] Ψ (P0) P0(b0(W) = 0) P0(b0(W) < 0)
N1 W1 + W2 0 l 0
N2 −0.2A[(W1z0.8)+]2 0 0.80 0.20
N3 −0.25A 0 0 l
A1 0.8A 0.19 0 0
A2 AW1+AW2+ 0.06 0.25 0.38
A3 AW1 0.09 0 0.50

Note: Decimals rounded to the nearest hundredth. For N2, z0.8 ≈ 0.84 is the 80th percentile of a standard normal distribution. We use x+ to denote the positive part of a real number x.