awoldemichael / TestPrivacyUnbiased

Corrects coefficient estimates from OLS regression when data matrix contains known measurement error

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

PrivacyUnbiased

This software implements the methods described in Georgina Evans and Gary King, "Statistically Valid Inferences from Differentially Private Data Releases"; See http://garyking.org/dpd

Install the software by running the following code:

devtools::install_github("georgieevans/PrivacyUnbiased")

library(PrivacyUnbiased)

Read an overview of the package:

https://github.com/georgieevans/PrivacyUnbiased/blob/master/Example/Example.pdf

PrivacyUnbiased provides methods to analyze differentially private data. For a system to release differentially private statistics from privacy protected data as presented in https://gking.harvard.edu/files/gking/files/udp.pdf, see our software UnbiasedPrivacy:

https://github.com/schwenzfeier/udp

About

Corrects coefficient estimates from OLS regression when data matrix contains known measurement error


Languages

Language:R 98.3%Language:TeX 1.7%