Warning when imputing level 2 factor variables with 2lonly.pmm

[reiniervlinschoten]

A weird bug is occuring, maybe related to #410.
When imputing a factor variable that should be constant within a class (e.g. smoking for a patient with longitudinal measurements), something goes wrong.

Reprex (from the vignette of the 2lonly.pmm function) which gives as a warning:
Warning message:

In `[<-.factor`(`*tmp*`, cc, value = c(`64` = 1, `64` = 1, `64` = 1,  :
  invalid factor level, NA generated

And the resulting dataframe has missing values in column V.

# simulate some data
# x,y ... level 1 variables
# v,w ... level 2 variables

G <- 250 # number of groups
n <- 20 # number of persons
beta <- .3 # regression coefficient
rho <- .30 # residual intraclass correlation
rho.miss <- .10 # correlation with missing response
missrate <- .50 # missing proportion
y1 <- rep(rnorm(G, sd = sqrt(rho)), each = n) + rnorm(G * n, sd = sqrt(1 - rho))
w <- rep(round(rnorm(G), 2), each = n)
v <- rep(round(runif(G, 0, 3)), each = n)
x <- rnorm(G * n)
y <- y1 + beta * x + .2 * w + .1 * v
dfr0 <- dfr <- data.frame("group" = rep(1:G, each = n), "x" = x, "y" = y, "w" = w, "v" = v)
dfr[rho.miss * x + rnorm(G * n, sd = sqrt(1 - rho.miss)) < qnorm(missrate), "y"] <- NA
dfr[rep(rnorm(G), each = n) < qnorm(missrate), "w"] <- NA
dfr[rep(rnorm(G), each = n) < qnorm(missrate), "v"] <- NA

# empty mice imputation
imp0 <- mice(as.matrix(dfr), maxit = 0)
predM <- imp0$predictorMatrix
impM <- imp0$method

# multilevel imputation
predM1 <- predM
predM1[c("w", "y", "v"), "group"] <- -2
predM1["y", "x"] <- 1 # fixed x effects imputation
impM1 <- impM
impM1[c("y", "w", "v")] <- c("2l.pan", "2lonly.norm", "2lonly.pmm")

# turn v into a categorical variable
dfr$v <- as.factor(dfr$v)
levels(dfr$v) <- LETTERS[1:4]

# y ... imputation using pan
# w ... imputation at level 2 using norm
# v ... imputation at level 2 using pmm

# skip imputation on solaris
is.solaris <- function() grepl("SunOS", Sys.info()["sysname"])
if (!is.solaris()) {
  imp <- mice(dfr,
    m = 1, predictorMatrix = predM1,
    method = impM1, maxit = 1, paniter = 500
  )
}

[hanneoberman]

Hi! A quick fix could be to convert the factor to a numeric variable before imputation, and recoding back afterwards?

[stefvanbuuren]

Thanks.

I patched mice.impute.2lonly.pmm() in mice 3.15.4, so we should not see the warning anymore.