Table 4 Bias in Scenarios with λ = 2, Outcome Prevalence = 0.30 and β 1 = β 2 = log(3)
From: Multiple imputation for handling missing outcome data when estimating the relative risk
Simulation scenario | Parameter | Coordinated missing data in Y and X 2 (FCS) | Missing data in Y only | Missing data in X 2 only |
|---|---|---|---|---|
Binary X 1 and X 2, RR(X 1, X 2) = 2 | β 1 | 0.03 | 0.02 | 0.02 |
β 2 | −0.11 | −0.06 | −0.05 | |
Continuous X 1 and X 2, Corr(X 1, X 2) = 0.70 | β 1 | −0.09 | −0.08 | −0.03 |
β 2 | −0.24 | −0.07 | −0.22 |