MaskJointDensity (version 1.0)

mask: Core function the Data Provider uses to mask the confidential original data to give to the user

Description

Use the multiplicative noise method to mask micro data and creates a vector of masked data, and creates a noise file as well.

Usage

mask(vectorToBeMasked, noisefile, noise, 
lowerBoundAsGivenByProvider=min(vectorToBeMasked),
upperBoundAsGivenByProvider=max(vectorToBeMasked),
maxorder=100,EPS=1e-06)

Arguments

vectorToBeMasked

data vector. All entries must be numeric or categorical. Missing values (NAs) are not allowed. If vectorToBeMasked are categorical, the argument is given by factor(vectorToBeMasked).

noisefile

a binary output file. The file name is ended by .bin. This file contains the information of lowerBoundAsGivenByProvider, upperBoundAsGivenByProvider, maxorder, the levels of the categorical data if vectorToBeMasked is categorical and EPS.

noise

noise data used to mask vectorToBeMasked. The size of the noise data is the same as the size of vectorToBeMasked.

lowerBoundAsGivenByProvider

The lower boundary used in evaluating the estimated density approximant. The default value is min(vectorToBeMasked). To protect min(vectorToBeMasked) and achieve an accurate estimate of the density function of vectorToBeMasked, the value of lowerBoundAsGivenByProvider is critical. In any case, the value of lowerBoundAsGivenByProvider should be less than min(vectorToBeMasked).

upperBoundAsGivenByProvider

The upper boundary used in evaluating the estimated density approximant. The default value is max(vectorToBeMasked). To protect max(vectorToBeMasked) and achieve an accurate estimate of the density function of vectorToBeMasked, the value of upperBoundAsGivenByProvider is critical. In any case, the value of upperBoundAsGivenByProvider should be greater than max(vectorToBeMasked).

maxorder

the maximum order of the moments in the sample-moment-based density approximate to be tested. The default value is 100

EPS

a threshold value. The default value is 1e-06.

Value

Returns a list with two elements.

ystar

masked data

noisefile

the title of a file containing the information of noise

Details

This R function is used to mask micro data by the multiplicative noise method and to produce a binary noise file for R function unmask. This file contains a sample of noise and other relevant information required by R function unmask. The size of the sample of noise stored in the file is ten times the size of vectorToBeMasked.

It is up to the user of mask to write the masked data to a file to provide to the end user.

References

Lin, Yan-Xia and Fielding, Mark James (2015). MaskDensity14: An R Package for the Density Approximant of a Univariate Based on Noise Multiplied Data, SoftwareX (accepted)

Examples

# NOT RUN {
##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--	or do  help(data=index)  for the standard data sets.

set.seed(123)
n=10000
y <- rmulti(n=10000, mean=c(30, 50), sd=c(4,2), p=c(0.3, 0.7))
      # y is a sample drawn from Y.
noise<-rmulti(n=10000, mean=c(80, 100), sd=c(5,3), p=c(0.6, 0.4))
      # noise is a sample drawn from C.
a1<-runif(1, min=min(y)-2,max=min(y))
b1<-runif(1, min=max(y), max=max(y)+2)
ymask<-mask(vectorToBeMasked = y, noisefile=file.path(tempdir(),"noise.bin"), noise,
lowerBoundAsGivenByProvider=a1, upperBoundAsGivenByProvider=b1)
write(ymask$ystar, file.path(tempdir(),"ystar.dat"))
# }