replicated.uniques: Replications in synthetic data

In synthpop: Generating Synthetic Versions of Sensitive Microdata for Statistical Disclosure Control

Replications in synthetic data

Description

Determines which unique units in the synthesised data set(s) have combinations of variables in the keys as follows:

1) unique in original data

2) unique in the synthetic data set(s)

3) unique in synthetic data and present,but not necessarily unique in original

4) unique in synthetic and unique in original.

For each of 3) and 4) results are returned that identify the rows in the synthetic data with each type of unique. This function is called by sdc where there are options to include each type of unique.

Usage

replicated.uniques(object, data, keys = names(data))

## S3 method for class 'repuniq.synds'
print(x, ...)

Arguments

object	an object of class synds, which stands for 'synthesised data set'. It is typically created by function syn() and it includes object$m synthesised data set(s).
data	the original observed data set.
keys	Variables to be used as quasi-identifiers to check for unique combinations.
...	additional parameters
x	an object of class repuniq.synds; a result of a call to replicated.uniques().

Value

A list of class "repuniq.synds" with the following components:

m	number of synthetic data sets in object object$m
n	number of rows in data object$n
k	number of rows in of synthetic data set(s) in object object$k
res_tab	Table or list of tables with numbers and percentages of uniques
synU.rm	A vector of length object$k TRUE/FALSE values wherea TRUE value identifies a unique in synthetic and prtesent in the original
repU.rm	A vector of length object$k TRUE/FALSE values where a TRUE value identifies a replicated unique

See Also

sdc

Examples

ods <- SD2011[1:1000,c("sex","age","region","edu","marital","smoke")]
s1 <- syn(ods, m = 2)
replicated.uniques(s1,ods, keys = c("sex","age","region"))

synthpop documentation built on April 12, 2025, 1:36 a.m.