grandMean: Create a "grand mean" data set from a series of imputed data...

In ppanko/immapTools: Convenience tools for IMMAP-related analyses

Description

Create a "grand mean" data set from a series of imputed data sets

Usage

grandMean(
  stackedData,
  idName,
  contNames = NULL,
  discNames = NULL,
  dropNames = NULL
)

Arguments

stackedData	a data.frame of several stacked imputed data sets
idName	a character vector of length 1 providing the name of the imputation id variable.
contNames	a character vector providing the names of the continuous variables. Default is NULL.
discNames	a character vector providing the names of the discrete variables. Default is NULL.
dropNames	a character vector providing the names of the variables to be excluded from the operation. Default is NULL.

Details

This function is an extension of the aggregate function for imputed data sets that contain multiple variable types. The default behavior is to aggregate the continuous variables inferred from contNames using the arithmetic mean, and to aggregate the discrete variables inferred from discNames using mode, while the variables inferred from dropNames are removed.

Value

a data.frame with the number of columns equal to stackedData and the number of rows equal to the unique values in the id column.

See Also

aggregate

Examples

## Create example data
data(iris)
iris$id <- factor(1:nrow(iris))
irisStack <- do.call(rbind, rep(list(iris), 10))
irisStack$junk <- sample(1:2000, nrow(iris))

## Create name vectors
irisContNames <- names(which(sapply(iris, is.numeric)))
irisDiscNames <- "Species"
irisDropNames <- "junk"
irisIdName <- "id"

## Create grand mean
gm <- grandMean(              
  stackedData = irisStack,    
  idName      = irisIdName,   
  contNames   = irisContNames,
  discNames   = irisDiscNames,
  dropNames   = irisDropNames 
)                             

ppanko/immapTools documentation built on Nov. 21, 2019, 12:28 a.m.