dpHistogram-class: Differentially private histogram

In IQSS/PSI-Library: Differentially Private Statistical Releases for Privacy Preservation

Description

Differentially private histogram

Arguments

varType	Character, the variable type.
variable	Character, the variable name in the data frame.
n	Integer, the number of observations.
epsilon	Numeric, the privacy loss parameter.
accuracy	Numeric, the desired accuracy of the query.
rng	Numeric, a priori estimate of the lower and upper bounds of a variable taking numeric values. Ignored for categorical types.
bins	Character, the available bins or levels of a categorical variable.
nBins	Integer, the number of bins to release.
granularity	Numeric, the width of each histogram bin (i.e. the inverse of 'nBins'). Used to calculate histogram bins in comination with 'rng'.
alpha	Numeric, level of statistical significance, default 0.05.
delta	Numeric, probability of privacy loss beyond epsilon.
imputeRng	Numeric, a 2-tuple indicating the lower and upper bounds of the range from which NA values in numeric or integer-type variables should be imputed
imputeBins	Character (or numeric for logical variables), a list of bins from which NA values values in character-type variables should be imputed
impute	Boolean, a boolean value indicating if logical-type variables should have NA values imputed or not. If true, a logical variable histogram will have 2 bins, 0 and 1. If false, the histogram will have 3 bins: 0, 1, and NA.
nBoot	Numeric, the number of bootstrap iterations to do for bootstrapping (not used for version 1 release)
mechanism	Character, one of 'mechanismLaplace', 'mechanismStability'. May be NULL, in which mechanism will be chosen automatically.

IQSS/PSI-Library documentation built on Feb. 15, 2020, 9:03 p.m.