simsem (version 0.2-8)

simMissing: Construct a SimMissing object to create data with missingness and analyze missing data.

Description

Function creates a SimMissing object that can be passed to simResult for creating and analyzing data with missingness.

Usage

simMissing(cov=0, pmMCAR=0, pmMAR=0, nforms=0, itemGroups=list(0), timePoints=1, twoMethod=0, prAttr=0, numImps=0, ignoreCols=0, threshold=0, covAsAux=TRUE, logical=new("NullMatrix"), ...)

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.

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)

twoMethod

Vector of (percent missing, column index). Will put a given percent missing on that column in the matrix to simulate a two method planned missing data research design.

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.

timePoints

Number of timepoints items were measured over. For longitudinal data, planned missing designs will be implemented within each timepoint.

numImps

The number of imputations to be used when multiply imputing missing data. Setting numImps to 0 will use FIML to handle missing data.

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.

...

Additional arguments to be passed to amelia for imputation. Only used if numImps is greater than 0.

Value

• A simMissing object to be used with SimResult.

Details

Without specifying any any arguments, no missingness will be introduced. Covariates are required to impose MAR missing. Imputations will be performed with Amelia

See Also

• SimMissingfor the alternative way to save missing data feature for using in thesimResultfunction.
• runMIfor imputing missing data by multiple imputation and analyze the imputed data.

Examples

#Example of imposing 10% MCAR missing in all variables with no imputations (FIML method)
	Missing <- simMissing(pmMCAR=0.1)
	summary(Missing)
	
	loading <- matrix(0, 6, 1)
	loading[1:6, 1] <- NA
	LX <- simMatrix(loading, 0.7)
	RPH <- symMatrix(diag(1))
	RTD <- symMatrix(diag(6))
	CFA.Model <- simSetCFA(LY = LX, RPS = RPH, RTE = RTD)
	SimData <- simData(CFA.Model, 500)
	SimModel <- simModel(CFA.Model)
	
	#Create data
	dat <- run(SimData)
	
	#Impose missing
	dat <- run(Missing, dat)
	
	#Analyze data
	out <- run(SimModel, dat)
	summary(out)
	
	#Example to create simMissing object for 3 forms design at 3 timepoints with 10 imputations
	Missing <- simMissing(nforms=3, timePoints=3, numImps=10)