Package ‘synthpop’ September 4, 2020 Type Package Title Generating Synthetic Versions of Sensitive Microdata for Statistical Disclosure Control Version 1.6-0 Date 2020-09-03 Author Beata Nowok [aut, cre], Gillian M Raab [aut], Chris Dibben [ctb], Joshua Snoke [ctb], Caspar van Lissa [ctb] Maintainer Beata Nowok <
[email protected]> Description A tool for producing synthetic versions of microdata containing confidential informa- tion so that they are safe to be released to users for exploratory analysis. The key objective of gen- erating synthetic data is to replace sensitive original values with synthetic ones causing mini- mal distortion of the statistical information contained in the data set. Variables, which can be cat- egorical or continuous, are synthesised one-by-one using sequential modelling. Replace- ments are generated by drawing from conditional distributions fitted to the original data us- ing parametric or classification and regression trees models. Data are synthesised via the func- tion syn() which can be largely automated, if default settings are used, or with methods de- fined by the user. Optional parameters can be used to influence the disclosure risk and the analyt- ical quality of the synthesised data. For a description of the imple- mented method see Nowok, Raab and Dibben (2016) <doi:10.18637/jss.v074.i11>. License GPL-2 | GPL-3 URL https://www.synthpop.org.uk/ Imports lattice, MASS, methods, nnet, ggplot2, graphics, stats, utils, rpart, party, foreign, plyr, proto, polspline, randomForest, ranger, classInt, mipfp LazyData yes NeedsCompilation no Repository CRAN Date/Publication 2020-09-04 08:10:02 UTC 1 2 R topics documented: R topics documented: synthpop-package .