dataDSS {GJRM.data} | R Documentation |
Simulated data with double sample selection
Description
Simulated data with double sample selection and binary outcome.
Usage
data(dataDSS)
Format
dataDSS
is a 10000 row data frame with the following columns:
- y1
First selection.
- y2
Second selection.
- y3
Binary outcome.
- x1, x2
Covariates.
- x3
Covariate influencing only
y1
.- x4
Covariate influencing only
y2
.- y3.o
Original outcome, without missingness.
Examples
# Data have been simulated as shown below
n <- 10000
x1 <- round(runif(n))
x2 <- runif(n)
x3 <- runif(n)
x4 <- rnorm(n)
u <- rnorm(n)
y1 <- ifelse(-1.55 + x1 - x2 + x3 + u + rnorm(n) > 0, 1, 0)
y2 <- ifelse(-0.25 - 0.5*x1 + x2 + x4 + u + rnorm(n) > 0, 1, 0)
y3 <- y3.o <- ifelse( -0.75 + x1 + x2 + u + rnorm(n) > 0, 1, 0)
y2 <- y2*y1
y3 <- y3*y2
y3 <- ifelse(y2 == 0, NA, y3)
dataDSS <- data.frame(y1, y2, y3, x1, x2, x3, x4, y3.o)
[Package GJRM.data version 0.1-1 Index]