Compute covariance matrix of incomplete data using multiple imputation. For multiple imputation, Multivariate Imputation by Chained Equations (MICE) from the mice package is used. The covariance matrices of the imputed data sets are combined using Rubin's rules.
mifa(
data,
cov_vars = dplyr::everything(),
n_pc,
ci = FALSE,
conf = 0.95,
n_boot = 1000,
...
)
A data frame with missing values coded as NA
.
Variables in data
for which to calculate the covariance
matrix. Supports (tidy selection)dplyr::select()
. This allows to
select variables that are used for the imputations of missing values, but not
the calculations of the covariance matrix. This is especially useful when
there are categorical predictors that can improve the imputation of the
response variables, but for which covariance cannot be calculated.
By default, all variables in data
are used for both, the imputation and
the covariance matrix. Note: Variables and rows used for the imputation, as
well as the method for imputation can be configured using the ...
.
See also mice::mice()
.
Integer or integer vector indicating number of principal components (eigenvectors) for which explained variance (eigenvalues) should be obtained and for which confidence intervals should be computed. Defaults to all principal components, i.e., the number of variables in the data.
A character string indicating which types of confidence intervals
should be constructed for the variance explained by the principal
components. If "boot"
, "fieller"
, or "both"
, the corresponding
intervals are computed. If FALSE
(the default) no confidence intervals will
be computed. The components for which confidence intervals should be computed
can be set with n_pc
. See mifa_ci_boot()
and mifa_ci_fieller()
for
details about the two methods.
Confidence level for constructing confidence intervals. The
default is .95
that is, 95% confidence intervals.
Number of bootstrap samples to use for bootstrapped confidence intervals. The default is 1000.
Arguments passed on to mice::mice
m
Number of multiple imputations. The default is m=5
.
method
Can be either a single string, or a vector of strings with
length length(blocks)
, specifying the imputation method to be
used for each column in data. If specified as a single string, the same
method will be used for all blocks. The default imputation method (when no
argument is specified) depends on the measurement level of the target column,
as regulated by the defaultMethod
argument. Columns that need
not be imputed have the empty method ""
. See details.
predictorMatrix
A numeric matrix of length(blocks)
rows
and ncol(data)
columns, containing 0/1 data specifying
the set of predictors to be used for each target column.
Each row corresponds to a variable block, i.e., a set of variables
to be imputed. A value of 1
means that the column
variable is used as a predictor for the target block (in the rows).
By default, the predictorMatrix
is a square matrix of ncol(data)
rows and columns with all 1's, except for the diagonal.
Note: For two-level imputation models (which have "2l"
in their names)
other codes (e.g, 2
or -2
) are also allowed.
ignore
A logical vector of nrow(data)
elements indicating
which rows are ignored when creating the imputation model. The default
NULL
includes all rows that have an observed value of the variable
to imputed. Rows with ignore
set to TRUE
do not influence the
parameters of the imputation model, but are still imputed. We may use the
ignore
argument to split data
into a training set (on which the
imputation model is built) and a test set (that does not influence the
imputation model estimates).
Note: Multivariate imputation methods, like mice.impute.jomoImpute()
or mice.impute.panImpute()
, do not honour the ignore
argument.
where
A data frame or matrix with logicals of the same dimensions
as data
indicating where in the data the imputations should be
created. The default, where = is.na(data)
, specifies that the
missing data should be imputed. The where
argument may be used to
overimpute observed data, or to skip imputations for selected missing values.
Note: Imputation methods that generate imptutations outside of
mice
, like mice.impute.panImpute()
may depend on a complete
predictor space. In that case, a custom where
matrix can not be
specified.
blocks
List of vectors with variable names per block. List elements
may be named to identify blocks. Variables within a block are
imputed by a multivariate imputation method
(see method
argument). By default each variable is placed
into its own block, which is effectively
fully conditional specification (FCS) by univariate models
(variable-by-variable imputation). Only variables whose names appear in
blocks
are imputed. The relevant columns in the where
matrix are set to FALSE
of variables that are not block members.
A variable may appear in multiple blocks. In that case, it is
effectively re-imputed each time that it is visited.
visitSequence
A vector of block names of arbitrary length, specifying the
sequence of blocks that are imputed during one iteration of the Gibbs
sampler. A block is a collection of variables. All variables that are
members of the same block are imputed
when the block is visited. A variable that is a member of multiple blocks
is re-imputed within the same iteration.
The default visitSequence = "roman"
visits the blocks (left to right)
in the order in which they appear in blocks
.
One may also use one of the following keywords: "arabic"
(right to left), "monotone"
(ordered low to high proportion
of missing data) and "revmonotone"
(reverse of monotone).
Special case: If you specify both visitSequence = "monotone"
and
maxit = 1
, then the procedure will edit the predictorMatrix
to conform to the monotone pattern. Realize that convergence in one
iteration is only guaranteed if the missing data pattern is actually
monotone. The procedure does not check this.
formulas
A named list of formula's, or expressions that
can be converted into formula's by as.formula
. List elements
correspond to blocks. The block to which the list element applies is
identified by its name, so list names must correspond to block names.
The formulas
argument is an alternative to the
predictorMatrix
argument that allows for more flexibility in
specifying imputation models, e.g., for specifying interaction terms.
blots
A named list
of alist
's that can be used
to pass down arguments to lower level imputation function. The entries
of element blots[[blockname]]
are passed down to the function
called for block blockname
.
post
A vector of strings with length ncol(data)
specifying
expressions as strings. Each string is parsed and
executed within the sampler()
function to post-process
imputed values during the iterations.
The default is a vector of empty strings, indicating no post-processing.
Multivariate (block) imputation methods ignore the post
parameter.
defaultMethod
A vector of length 4 containing the default
imputation methods for 1) numeric data, 2) factor data with 2 levels, 3)
factor data with > 2 unordered levels, and 4) factor data with > 2
ordered levels. By default, the method uses
pmm
, predictive mean matching (numeric data) logreg
, logistic
regression imputation (binary data, factor with 2 levels) polyreg
,
polytomous regression imputation for unordered categorical data (factor > 2
levels) polr
, proportional odds model for (ordered, > 2 levels).
maxit
A scalar giving the number of iterations. The default is 5.
printFlag
If TRUE
, mice
will print history on console.
Use print=FALSE
for silent computation.
seed
An integer that is used as argument by the set.seed()
for
offsetting the random number generator. Default is to leave the random number
generator alone.
data.init
A data frame of the same size and type as data
,
without missing data, used to initialize imputations before the start of the
iterative process. The default NULL
implies that starting imputation
are created by a simple random draw from the data. Note that specification of
data.init
will start all m
Gibbs sampling streams from the same
imputation.
A list:
The estimated covariance matrix of the incomplete data, based on the combined covariance matrices of imputed data sets.
A list containing the estimated covariance matrixes for all imputed data sets.
A data frame containing the estimated proportions of
explained variance for each of specified n_pc
components. Depending o
n ci
, it will also contain the estimated Fieller's (parametric) and/or
bootstrap (nonparametric) confidence interval for the proportion of
variance explained by the different numbers of principal components defined
by n_pc
.
Object of type mice::mids. This is the results of the multiple imputation step for the covariance matrix. Can be useful for diagnosing the multiple imputations.
The function also computes the variance explained by different numbers of principal components and the corresponding Fieller (parametric) or bootstrap (nonparametric) confidence intervals.
Nassiri, V., Lovik, A., Molenberghs, G., & Verbeke, G. (2018). On using multiple imputation for exploratory factor analysis of incomplete data. Behavioral Research Methods 50, 501–517. doi:10.3758/s13428-017-1013-4
# \donttest{
if(requireNamespace("psych")) {
data <- psych::bfi
mifa(data, cov_vars = -c(age, education, gender), ci = "fieller", print = FALSE)
}
#> Imputed covariance matrix of 25 variables
#>
#> Variable: A1 A2 A3 A4 A5 C1 C2 C3 C4 C5 E1 E2 E3 E4 E5 N1 N2 N3 N4 N5 O1 O2 O3 O4 O5
#> N Imputed: 16 27 26 19 16 21 24 20 26 16 23 16 25 9 21 22 21 11 36 29 22 0 28 14 20
#>
#> Number of MICE imputations: 5
#> Additional variables used for imputations:
#> gender education age
#>
#> Cumulative proportion of variance explained by n principal components:
#>
#> n prop Fieller CI
#> 1 0.21 [0.20, 0.22]
#> 2 0.33 [0.32, 0.34]
#> 3 0.41 [0.40, 0.42]
#> 4 0.48 [0.47, 0.49]
#> 5 0.54 [0.53, 0.55]
#> 6 0.58 [0.58, 0.59]
#> 7 0.62 [0.61, 0.63]
#> 8 0.66 [0.65, 0.66]
#> 9 0.69 [0.68, 0.69]
#> 10 0.72 [0.71, 0.72]
#> 11 0.74 [0.74, 0.75]
#> 12 0.77 [0.76, 0.77]
#> 13 0.79 [0.79, 0.79]
#> 14 0.81 [0.81, 0.82]
#> 15 0.83 [0.83, 0.84]
#> 16 0.85 [0.85, 0.86]
#> 17 0.87 [0.87, 0.88]
#> 18 0.89 [0.89, 0.90]
#> 19 0.91 [0.91, 0.91]
#> 20 0.93 [0.93, 0.93]
#> 21 0.94 [0.94, 0.95]
#> 22 0.96 [0.96, 0.96]
#> 23 0.97 [0.97, 0.97]
#> 24 0.99 [0.99, 0.99]
#> 25 1.00 [1.00, 1.00]
#>
# }