miceadds (version 3.17-44)

mice.impute.rlm: Imputation of a Linear Model by Bayesian Bootstrap

Description

These functions impute from linear models using the functions stats::lm, MASS::rlm or MASS::lqs. The method mice.impute.lm_fun allows the definition of a general linear regression fitting function for which the methods predict and residuals are defined.

Parameters of the model are estimated by Bayesian bootstrap. Predicted values are computed and residuals are randomly drawn from the empirical distribution of residuals of observed data.

Usage

mice.impute.lm(y, ry, x, wy=NULL, lm_args=NULL, trafo=NULL, antitrafo=NULL, ...)
mice.impute.rlm(y, ry, x, wy=NULL, lm_args=NULL, trafo=NULL, antitrafo=NULL, ...)
mice.impute.lqs(y, ry, x, wy=NULL, lm_args=NULL, trafo=NULL, antitrafo=NULL, ...)
mice.impute.lm_fun(y, ry, x, wy=NULL, lm_args=NULL, lm_fun="lm", trafo=NULL,
               antitrafo=NULL, ...)

Value

A vector of length nmis=sum(!ry) with imputed values.

Arguments

y

Incomplete data vector of length n

ry

Vector of missing data pattern (FALSE -- missing, TRUE -- observed)

x

Matrix (n x p) of complete covariates.

wy

Vector of logicals indicating which entries should be imputed

lm_args

List of arguments for stats::lm, MASS::rlm, MASS::lqs or a user-defined function.

lm_fun

Linear regression fitting function, e.g. stats::lm for which S3 methods predict and residuals are defined.

trafo

Optional function for transforming the outcome values

antitrafo

Optional function which is the inverse function of trafo

...

Further arguments to be passed

Examples

if (FALSE) {
#############################################################################
# EXAMPLE 1: Some toy example illustrating the methods
#############################################################################

library(MASS)
library(mice)

#-- simulate data
set.seed(98)
N <- 1000
x <- stats::rnorm(N)
z <- 0.5*x + stats::rnorm(N, sd=.7)
y <- stats::rnorm(N, mean=.3*x - .2*z, sd=1 )
dat <- data.frame(x,z,y)
dat[ seq(1,N,3), c("x","y") ] <- NA
dat[ seq(1,N,4), "z" ] <- NA

#-- define imputation methods
imp <- mice::mice(dat, maxit=0)
method <- imp$method
method["x"] <- "rlm"
method["z"] <- "lm"
method["y"] <- "lqs"

#-- impute data
imp <- mice::mice(dat, method=method)
summary(imp)

#--- example using transformations
dat1$x <- exp(dat1$x)
dat1$z <- stats::plogis(dat1$z)

trafo <- list(x=log, z=stats::qlogis)
antitrafo <- list(x=exp, z=stats::plogis)

#- impute with transformations
imp2 <- mice::mice(dat1, method=method, m=1, maxit=3, trafo=trafo, antitrafo=antitrafo)
print(imp2)
}