miss {simsem} | R Documentation |
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,
package = "default", convergentCutoff = 0.8, ignoreCols = 0,
threshold = 0, covAsAux = TRUE, logical = NULL, ...)
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 |
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 |
m |
The number of imputations. The default is 0 such that the full information maximum likelihood is used. |
package |
The package to be used in multiple imputation. The default value of this function is |
convergentCutoff |
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. |
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 |
logical |
A matrix of logical values ( |
... |
Additional arguments used in multiple imputation function. |
Value
A missing object that contains missing-data template (SimMissing
)
Author(s)
Alexander M. Schoemann (East Carolina University; schoemanna@ecu.edu), Patrick Miller (University of Notre Dame; pmille13@nd.edu), Sunthud Pornprasertmanit (psunthud@gmail.com)
See Also
-
SimMissing
The resulting missing object
Examples
#Example of imposing 10% MCAR missing in all variables with no imputations (FIML method)
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 with 10 imputations
Missing <- miss(nforms=3, timePoints=3, numImps=10)
#Missing template for data analysis with multiple imputation
Missing <- miss(package="mice", m=10, convergentCutoff=0.6)