README.md

In bonorico/gcipdr: Gaussian Copula (based) Individual Person Data (IPD) Reconstruction

gcipdr

Gaussian Copula Individual Person Data Reproduction

Installation instructions in R (follow in order):

step 1: open your R and run command below

library(devtools)
install_github("bonorico/gcipdr")

step 2: Important !

gcipdr depends on CRAN archived package 'JohnsonDistribution'. You must install this dependency from source by doing the following:

 url <- "https://cran.r-project.org/src/contrib/Archive/JohnsonDistribution/JohnsonDistribution_0.24.tar.gz"
 pkgFile <- "JohnsonDistribution_0.24.tar.gz"
 download.file(url = url, destfile = pkgFile)

 install.packages(pkgs=pkgFile, type="source", repos=NULL)

 unlink(pkgFile)

step 3: getting help

help("gcipdr")

~+~+~+~+~+~+~+~ ~+~+~+~+~+~+~+~ ~+~+~+~+~+~+~+~ ~+~+~+~+~+~+~+~ ~+~+~+~+~+~+~+~ ~+~+~+~+~+~+~+~ ~+~+~+~+~+~+~+~

Package Description

The proposed Gaussian Copula technique generates pseudodata replacing IPD, using simple summaries, like IPD empirical marginal moments and correlation matrix, as only input data. In contexts where IPD is not available, pseudodata is a mean to compute approximate IPD inferences. The approach has applications in statistical disclosure control, distributed computing (e.g. see DataShield -- see https://cran.datashield.org/), and, ideally research synthesis or reproduction.

Further Details

This package (documentation) is yet to receive furhter editing.

HOW TO CITE

Cite this software as follows:

Bonofiglio Federico. (2018). The gcipdr package version 0.0: Gaussian copula (based) individual person data reproduction. URL: https://github.com/bonorico/gcipdr.

Bibtex:

@misc{Bono_github_gcipdr,
  author = {Bonofiglio, Federico},
  title = {The {\texttt gcipdr} package Version 0.0: Gaussian Copula (based) Individual Person Data Reproduction},
  year = {2018},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {URL: https://github.com/bonorico/gcipdr}
}

Installation Notes:

Executing command 'install_github()' above should cause to fresh (re)install LinkingTo packages listed in the DESCRIPTION file (Rcpp, RcppArmadillo). In my case the command also caused to fresh (re)install other dependencies ('cubature' and 'ggplot2') but not all of them. The full list of dependencies ('moment', 'parallel', 'cubature', 'mvtnorm', 'ggplot2') is issued in a warning at installation time. Install any missing dependency manually.

If you want to avoid automatically fresh re-installing any of the above dependencies, use the following ALTERNATIVE INSTALLATION PLAN (Linux commands -- should work similarly with other OS):

1) Download all content of the gcipdr repository on your local machine (in a folder called 'gcipdr')

2) From Terminal: cd into 'your_local_directory/gcipdr' and run:

                                                              R CMD build gcipdr

3) From R:

                                 pkg <- "your_local_directory/gcipdr_0.0.tar.gz"
                        install.packages(pkgs=pkg, type="source", repos=NULL)

4) From R: all remnant dependencies must be installed manually, if yet not installed on your machine.

Installation works fine on my local machine: Linux 3.13.0-147-generic #196-Ubuntu x86_64 (with 4 Intel CPUs) GNU/Linux. At installation you could receive the following

WARNING: support for OpenMP requires C++11/C++14; add -std=c++11 or -std=c++14 to compiler flags

You should be able to just ignore it.

Help files

After installation type '?DataRebuild' or '?Simulate.many.datasets' in the R command line to obtain more information on gcipdr modules.

bonorico/gcipdr documentation built on May 2, 2021, 8:12 p.m.