Learn R Programming

CDF.PSIdekick (version 1.2)

Evaluate Differentially Private Algorithms for Publishing Cumulative Distribution Functions

Description

Designed by and for the community of differential privacy algorithm developers. It can be used to empirically evaluate and visualize Cumulative Distribution Functions incorporating noise that satisfies differential privacy, with numerous options made to streamline collection of utility measurements across variations of key parameters, such as epsilon, domain size, sample size, data shape, etc. Developed by researchers at Harvard PSI.

Copy Link

Version

Install

install.packages('CDF.PSIdekick')

Monthly Downloads

14

Version

1.2

License

GPL (>= 2)

Maintainer

Daniel Muise

Last Published

August 19th, 2016

Functions in CDF.PSIdekick (1.2)

Abbrev

Tranforms long numbers into short strings.
diffatQuantile

Determine the distance between CDFs at key quantiles.
CDFtestTrack

Test a single CDF implementation with one set of parameters.
DerivDiff

Determine how well a single DPCDF matches the shape of its data.
CDFtestTrackx

Test a single CDF implementation with one set of parameters.
badCDF

Make a straight-line faux CDF.
diffat25

Determine the distance between CDFs at the .25 quantile.
CDFtest

Comprehensively evaluate and visualize the utility of CDF-generating implementations.
diffat75

Determine the distance between CDFs at the .75 quantile.
diffatMedian

Determine the distance between CDFs at the median.
functionH

Create a DP-CDF by creating a K-degree noisy tree
findMaxError

Locate where the maximum error occurs between two CDFs
dpCDFtesting-package

Comprehensively evaluate and visualize the output of dpCDF-generating algorithm implementations. dp = Differential Privacy
horzdiffat25

Determine the distance between the .25 quantile values returned by two CDFs.
functionSUB

Build dpCDFs through use of a noisy tree with bin merging.
horzdiffat75

Determine the distance between the .75 quantile values returned by two CDFs.
getMaxError

Determine an approximate CDF's maximum error.
getMean

Calculate the private mean from the DP-CDF
functionHmono

Create a monotonically increasing DP-CDF by creating a K-degree noisy tree
functionS2

Build dpCDFs through Histogram smoothing and minimized expected L2 per bin
KurtDiffpdf

Error in Kurtosis from CDF (under development)
MaxErrorAt_PDF

Locate where the maximum error occurs between two PDFs
MaxErrorAt_CDF

Locate where the maximum error occurs between two CDFs
L1empiric

Calculate the area between two CDFs.
horzdiffatQuantile

Determine the distance between the quantile values returned by two CDFs.
horzdiffatMed

Determine the distance between the median values returned by two CDFs.
MaxError_CDF

Determine an approximate CDF's maximum error.
MaxError_PDF

Determine an approximate PDF's maximum error.
L2empiric

Calculate the empirical L2norm between two CDFs.
MAE

Calculate the MAE of a dpCDF relative to that of the non-private CDF.
SDempiric

Calculate the std. dev. on a DPCDF.
ModeDiffpdf

Error in Mode from CDF
MSEanalytic

Determine the expected MSE of a simple DPCDF from its parameters.
MeanDiffpdf

Error in mean from CDF
MovetoRange

Clamp a value to a specified range.
SkewDiffpdf

Error in Skewness from CDF (under development)
MSE

Calculate the MSE of a DP-CDF relative to the non-private CDF.
Medians

Retrieve a median estimate from the dpCDF
nodes

Node parser.
QuantileFromCDF

Retrieve a private quantile estimate from the dpCDF
VarDiffpdf

Error in Variance from CDF
TreeCDF

Creates a Tree then a CDF
StdDiffpdf

Error in Standard Deviation from CDF
smoothVector2

Enforce monotnocity on a vector.
Smooth

Monotonicity enforcement