Missing function

Missing value generator

Missing(object, formula, Rformula, missing.name, suffix = "0", ...)

Arguments

• object: lvm-object.
• formula: The right hand side specifies the name of a latent variable which is not always observed. The left hand side specifies the name of a new variable which is equal to the latent variable but has missing values. If given as a string then this is used as the name of the latent (full-data) name, and the observed data name is 'missing.data'
• Rformula: Missing data mechanism with left hand side specifying the name of the observed data indicator (may also just be given as a character instead of a formula)
• missing.name: Name of observed data variable (only used if 'formula' was given as a character specifying the name of the full-data variable)
• suffix: If missing.name is missing, then the name of the oberved data variable will be the name of the full-data variable + the suffix
• ...: Passed to binomial.lvm.

Returns

lvm object

Details

This function adds a binary variable to a given lvm model and also a variable which is equal to the original variable where the binary variable is equal to zero

Examples

library(lava)
set.seed(17)
m <- lvm(y0~x01+x02+x03)
m <- Missing(m,formula=x1~x01,Rformula=R1~0.3*x02+-0.7*x01,p=0.4)
sim(m,10)

m <- lvm(y~1)
m <- Missing(m,"y","r")
## same as
## m <- Missing(m,y~1,r~1)
sim(m,10)

## same as
m <- lvm(y~1)
Missing(m,"y") <- r~x
sim(m,10)

m <- lvm(y~1)
m <- Missing(m,"y","r",suffix=".")
## same as
## m <- Missing(m,"y","r",missing.name="y.")
## same as
## m <- Missing(m,y.~y,"r")
sim(m,10)

Author(s)

Thomas A. Gerds tag@biostat.ku.dk