Microaggregation for numerical and categorical key variables based on a distance similar to the Gower Distance

The microaggregation is based on the distances computed similar to the Gower distance. The distance function makes distinction between the variable types factor,ordered,numerical and mixed (semi-continuous variables with a fixed probability mass at a constant value e.g. 0)

microaggrGower(
  obj,
  variables = NULL,
  aggr = 3,
  dist_var = NULL,
  by = NULL,
  mixed = NULL,
  mixed.constant = NULL,
  trace = FALSE,
  weights = NULL,
  numFun = mean,
  catFun = VIM::sampleCat,
  addRandom = FALSE
)

Arguments

obj

sdcMicroObj-class-object or a data.frame

variables

character vector with names of variables to be aggregated (Default for sdcMicroObj is all keyVariables and all numeric key variables)

aggr

aggregation level (default=3)

dist_var

character vector with variable names for distance computation

by

character vector with variable names to split the dataset before performing microaggregation (Default for sdcMicroObj is strataVar)

mixed

character vector with names of mixed variables

mixed.constant

numeric vector with length equal to mixed, where the mixed variables have the probability mass

trace

TRUE/FALSE for some console output

weights

numerical vector with length equal the number of variables for distance computation

numFun

function: to be used to aggregated numerical variables

catFun

function: to be used to aggregated categorical variables

addRandom

TRUE/FALSE if a random value should be added for the distance computation.

Value

The function returns the updated sdcMicroObj or simply an altered data frame.

Details

The function sampleCat samples with probabilities corresponding to the occurrence of the level in the NNs. The function maxCat chooses the level with the most occurrences and random if the maximum is not unique.

Note

In each by group all distance are computed, therefore introducing more by-groups significantly decreases the computation time and memory consumption.

See also

sampleCat and maxCat

Author

Alexander Kowarik

Examples

data(testdata,package="sdcMicro")
testdata <- testdata[1:200,]
# \donttest{
for(i in c(1:7,9)) testdata[,i] <- as.factor(testdata[,i])
test <- microaggrGower(testdata,variables=c("relat","age","expend"),
  dist_var=c("age","sex","income","savings"),by=c("urbrur","roof"))

sdc <- createSdcObj(testdata,
  keyVars=c('urbrur','roof','walls','water','electcon','relat','sex'),
  numVars=c('expend','income','savings'), w='sampling_weight')

sdc <- microaggrGower(sdc)
# }