mlmi (version 1.1.2)

normImp: Multivariate normal model imputation

Description

This function performs multiple imputation under a multivariate normal model as described by Schafer (1997), using his norm package, either with or without posterior draws.

Usage

normImp(obsData, M = 10, pd = FALSE, steps = 100, rseed)

Value

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

Arguments

obsData

The data frame to be imputed.

M

Number of imputations to generate.

pd

Specify whether to use posterior draws (TRUE) or not (FALSE).

steps

If pd is TRUE, the steps argument specifies how many MCMC iterations to perform.

rseed

The value to set the norm package's random number seed to, using the rngseed function of norm. This function must be called at least once before imputing using norm. If the user wishes to set the seed using rngseed before calling normImp, set rseed=NULL.

Details

This function imputes from a multivariate normal model with unstructured covariance matrix, as described by Schafer (1997). With pd=FALSE, all imputed datasets are generated conditional on the MLE of the model parameter, referred to as maximum likelihood multiple imputation by von Hippel and Bartlett (2021).

With pd=TRUE, regular 'proper' multiple imputation is used, where each imputation is drawn from a distinct value of the model parameter. Specifically, for each imputation, a single MCMC chain is run, iterating for steps iterations.

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

References

Schafer J.L. (1997). Analysis of incomplete multivariate data. Chapman & Hall, Boca Raton, Florida, USA.

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 tools:::Rd_expr_doi("10.1214/20-STS793").

Examples

#simulate a partially observed dataset from multivariate normal distribution
set.seed(1234)
n <- 100
temp <- MASS::mvrnorm(n=n,mu=rep(0,4),Sigma=diag(4))

#make some values missing
for (i in 1:4) {
  temp[(runif(n)<0.25),i] <- NA
}

#impute using normImp
imps <- normImp(data.frame(temp), M=10, pd=FALSE, rseed=4423)