normUniImp: Normal regression imputation of a single variable

In jwb133/mlmi: Maximum Likelihood Multiple Imputation

Normal regression imputation of a single variable

Description

Performs multiple imputation of a single continuous variable using a normal linear regression model. The covariates in the imputation model must be fully observed. By default normUniImp imputes every dataset using the maximum likelihood estimates of the imputation model parameters, which here coincides with the OLS estimates, referred to as maximum likelihood multiple imputation by von Hippel and Bartlett (2021). If pd=TRUE is specified, it instead performs posterior draw Bayesian imputation.

Usage

normUniImp(obsData, impFormula, M = 5, pd = FALSE)

Arguments

obsData	The data frame to be imputed.
impFormula	The linear model formula.
M	Number of imputations to generate.
pd	Specify whether to use posterior draws (TRUE) or not (FALSE).

Details

Imputed datasets can be analysed using withinBetween, scoreBased, or for example the bootImpute package.

Value

A list of imputed datasets, or if M=1, just the imputed data frame.

References

von Hippel P.T. and Bartlett J.W. Maximum likelihood multiple imputation: faster, more efficient imputation without posterior draws. Statistical Science 2021; 36(3) 400-420 \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1214/20-STS793")}.

Examples

#simulate a dataset with one partially observed (conditionally) normal variable
set.seed(1234)
n <- 100
x <- rnorm(n)
y <- x+rnorm(n)
x[runif(n)<0.25] <- NA
temp <- data.frame(x=x,y=y)

#impute using normImp
imps <- normUniImp(temp, y~x, M=10, pd=FALSE)

jwb133/mlmi documentation built on June 4, 2023, 9:39 a.m.