DatawMis: Data with monotone missingness and nonmonotone missingness.

In mcl868/causalinmisdata: Causal inference in missing data

Data with monotone missingness and nonmonotone missingness.

Description

The data by the name "DataSetFull" contain the data with full data. The data by the name "DataSetMonotone" contain the data with missing observation following a monotone pattern. The missigness follow coarsening at random. The data by the name "DataSetnonMonotone" contain the data with missing observation following a nonmonotone pattern.

Arguments

L0	Baseline measurement.
A0	The exposure after baseline measurement.
L1	Confounder for the two exposure A1 and A2.
A1	The exposure at time 1.
L2	Confounder for the exposure A2.
A2	The exposure at time 2.
Y	The outcome.

Author(s)

Thomas Maltesen [e-mail]

References

put references to the literature/web site here

Examples

## Not run: 
p<-function(x)exp(x)/(1+exp(x))

L0<-rnorm(10000)
A0<-1*(runif(10000,0,1)<=p(0.6*L0))

L1<--A0+0.2*L0-1*A0*L0+rnorm(10000)
A1<-1*(runif(10000,0,1)<=p(-1+1.6*A0+1.2*L1-0.8*L0-1.6*L1*A0))

L2<--A1+1*L1-A0+1.2*L0+rnorm(10000)
A2<-1*(runif(10000,0,1)<=p(1-0.8*L0+1.6*A0+1.2*L1+1.3*A1+0.5*L2+1.6*L1*A1))

Y<-2*L0+3*A0+1*L1+2*A1-2*L2+5*A2+L2*A2+rnorm(10000)

DataSet<-data.frame(L0, L1, L2, A0, A1, A2, Y)
rm(list=c("L0","A0","L1","A1","L2","A2","Y"))

model1 <- Y ~ L0 + A0 + L1 + A1 + L2 + A2 + L2*A2
model2 <- model1 ~ A1 + L1 + A0 + L0
model3 <- model2 ~ A0 + L0 + A0*L0

g.aipw.dicho(mmodels=c(model1,model2,model3),
             exposure=c("A0","A1","A2"),
             data=DataSet)

## End(Not run)

mcl868/causalinmisdata documentation built on March 5, 2024, 8:22 a.m.