miss: Specifying the missing template to impose on a dataset

Description

Specifying the missing template (SimMissing) to impose on a dataset. The template will be used in Monte Carlo simulation such that, in the sim function, datasets are created and imposed by missing values created by this template. See imposeMissing for further details of each argument.

Usage

miss(cov = 0, pmMCAR = 0, pmMAR = 0, logit = "", nforms = 0, itemGroups = list(),
     timePoints = 1, twoMethod = 0, prAttr = 0,
     m = 0, convergentCutoff = 0.8, ignoreCols = 0,
     threshold = 0, covAsAux = TRUE, logical = NULL, ...)

Value

A missing object that contains missing-data template (SimMissing)

Arguments

cov: Column indices of any normally distributed covariates used in the data set.
pmMCAR: Decimal percent of missingness to introduce completely at random on all variables.
pmMAR: Decimal percent of missingness to introduce using the listed covariates as predictors.
logit: The script used for imposing missing values by logistic regression. The script is similar to the specification of regression in lavaan such that each line begins with a dependent variable, then '~' is used as regression sign, and the formula of a linear combination of independent variable plus constant, such as y1 ~ 0.5 + 0.2*y2. '#' and '!' can be used as a comment (like lavaan). For the intercept, users may use 'p()' to specify the average proportion of missing, such as y1 ~ p(0.2) + 0.3*y2, which the average missing proportion of y1 is 0.2 and the missing of y1 depends on y2. Users may visualize the missing proportion from the logistic specification by the plotLogitMiss function.
nforms: The number of forms for planned missing data designs, not including the shared form.
itemGroups: List of lists of item groupings for planned missing data forms. Without this, items will be divided into groups sequentially (e.g. 1-3,4-6,7-9,10-12)
timePoints: Number of timepoints items were measured over. For longitudinal data, planned missing designs will be implemented within each timepoint.
twoMethod: With missing on one variable: vector of (column index, percent missing). Will put a given percent missing on that column in the matrix to simulate a two method planned missing data research design. With missing on two or more variables: list of (column indices, percent missing).
prAttr: Probability (or vector of probabilities) of an entire case being removed due to attrition at a given time point. See imposeMissing for further details.
m: The number of imputations (DEPRECATED). The only currently accepted value is 0, indicating to use full information maximum likelihood estimation.
convergentCutoff: DEPRECATED. If the proportion of convergent results across imputations are greater than the specified value (the default is 80%), the analysis on the dataset is considered as convergent. Otherwise, the analysis is considered as nonconvergent. This attribute is applied for multiple imputation only (currently not supported).
ignoreCols: The columns not imposed any missing values for any missing data patterns
threshold: The threshold of covariates that divide between the area to impose missing and the area not to impose missing. The default threshold is the mean of the covariate.
covAsAux: If TRUE, the covariate listed in the object will be used as auxiliary variables when putting in the model object. If FALSE, the covariate will be included in the analysis.
logical: A matrix of logical values (TRUE/FALSE). If a value in the dataset is corresponding to the TRUE in the logical matrix, the value will be missing.
...: DEPRECATED. Additional arguments passed to a multiple-imputation function (no longer supported).

Author

Alexander M. Schoemann (East Carolina University; schoemanna@ecu.edu), Patrick Miller (University of Notre Dame; pmille13@nd.edu), Sunthud Pornprasertmanit (psunthud@gmail.com)

Examples

Run this code

#Example of imposing 10% MCAR missing in all variables (FIML estimation)
Missing <- miss(pmMCAR=0.1, ignoreCols="group")
summary(Missing)

loading <- matrix(0, 6, 1)
loading[1:6, 1] <- NA
LY <- bind(loading, 0.7)
RPS <- binds(diag(1))
RTE <- binds(diag(6))
CFA.Model <- model(LY = LY, RPS = RPS, RTE = RTE, modelType="CFA")

#Create data
dat <- generate(CFA.Model, n = 20)

#Impose missing
datmiss <- impose(Missing, dat)

#Analyze data
out <- analyze(CFA.Model, datmiss)
summary(out)

#Missing using logistic regression
script <- 'y1 ~ 0.05 + 0.1*y2 + 0.3*y3
	y4 ~ -2 + 0.1*y4
	y5 ~ -0.5'
Missing2 <- miss(logit=script, pmMCAR=0.1, ignoreCols="group")
summary(Missing2)
datmiss2 <- impose(Missing2, dat)

#Missing using logistic regression (2)
script <- 'y1 ~ 0.05 + 0.5*y3
	y2 ~ p(0.2)
	y3 ~ p(0.1) + -1*y1
	y4 ~ p(0.3) + 0.2*y1 + -0.3*y2
	y5 ~ -0.5'
Missing2 <- miss(logit=script)
summary(Missing2)
datmiss2 <- impose(Missing2, dat)

#Example to create simMissing object for 3 forms design at 3 timepoints
Missing <- miss(nforms=3, timePoints=3,
                itemGroups = list(X = 1:3,
                                  A = 4:6,
                                  B = 7:9,
                                  D = 10:12))