Skip to main content

Table 1 Simulation parameters for irregular visits, confounder, treatment and longitudinal outcome

From: Estimation of marginal structural models under irregular visits and unmeasured confounder: calibrated inverse probability weights

Description

Simulation parameters

\(Y_{ij}:\) Longitudinal outcomea

\(\theta _0=9;\theta _1=-0.3;\theta _2=-0.1\); \(\theta _3=0.5\);

 

\(\theta _4=\theta _5=\theta _6=0.5\); (\(\theta _{7}=0.1\)); \(\sigma ^2_y=3\)

\(L_{ij}:\) Time-dependent confounder

\(\mu _0=0;\mu _1=0.1; \mu _2=0\); \(\mu _3=0\); \(\mu _4=0.1\)

\(V_{ij}:\) Irregular visitsb

\(\omega _0=2 [16];\omega _1=0.1;\omega _2=0.1; \omega _3=0.1\); \(\omega _4=0.1\)

\(A_{ij}:\) Time-dependent treatmenta

\(\alpha _0=0;\alpha _1=0.1\); \(\alpha _2=0.1\) ; \(\alpha _3=0.1\);

 

\(\alpha _4=0\); \(\alpha _5=0.3\); \(\alpha _6=0.3\); \(\alpha _7=0.6\); (\(\alpha _8=0.1\))

N :  number of subjects

100

t :  number of discrete time-points

10

r :  number of replicates

1000

  1. a(.) denotes the coefficient for unmeasured confounder
  2. b[.] denotes the coefficient for absence of irregular visits