miss: Specifying the missing template to impose on a dataset

Description Usage Arguments Value Author(s) See Also Examples

View source: R/miss.R

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

1
2
3
4
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 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. 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 "default". For the default option, if m is 0, the full information maximum likelihood is used. If m is greater than 0, the "mice" package is used. The possible inputs are "default", "Amelia", or "mice".

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 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 used in multiple imputation function.

Value

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

Author(s)

Alexander M. Schoemann (East Carolina University; [email protected].edu), Patrick Miller (University of Notre Dame; [email protected]), Sunthud Pornprasertmanit ([email protected])

See Also

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
#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)

simsem documentation built on June 3, 2018, 5:04 p.m.

Related to miss in simsem...