Artificial or simulated data are particularly relevant in tests and benchmarks for machine learning methods, in teaching for exercises and for setting up analysis workflows. They are relevant when real data may not be used for reasons of data protection, or when special distributions or effects should be present in the data to test certain machine learning methods. In this paper a generator for multivariate numerical data with arbitrary marginal distributions and – as far as possible – arbitrary correlations is presented. The data generator is implemented in the open source statistics software R. It can also be used for categorical variables, if data are generated separately for the corresponding characteristics of a categorical variable. Additionally, outliers can be integrated. The use of the data generator is demonstrated with a concrete example.
Osei-BrysonK.-M. and BarclayC., Knowledge Discovery Process and Methods to Enhance Organizational Performance, CRC Press, Boca Raton, 2015.
2.
PeiY. and ZaïaneO.R., Synthetic Data Generator for Clustering and Outlier Analysis Technical report, 2006.
3.
RachkovskijD.A. and KussulE.M., DataGen: a generator of datasets for evaluation of classification algorithms, in: Pattern Recognition Letters19(7) (1998), 537–544.
4.
BischlB.SchiffnerJ. and WeihsC., Benchmarking Classification Algorithms on High-Performance Computing Clusters, in: Studies in Classification, Data Analysis, and Knowledge Organization, Springer, Heidelberg, 2014.
5.
DemlerO.V.PencinaM.J. and D’AgostinoR.B. Sr., Impact of correlation on predictive ability of biomarkers, in: Statistics in Medicine32(24) (2013), 4196–4210.
6.
WangZ.MylesP. and TuckerA., Generating and Evaluating Synthetic UK Primary Care Data: Preserving Data Utility & Patient Privacy, in: 2019 IEEE 32nd International Symposium on Computer-Based Medical Systems (CBMS), 2019, pp. 126–131.
7.
ParkY. and GhoshJ., PeGS: perturbed gibbs samplers that generate privacy-compliant synthetic data, in: Transactions on Data Privacy7(3) (2014), 253–282.
8.
DrechslerJ., Using support vector machines for generating synthetic datasets, in: International Conference on Privacy in Statistical Databases, Springer, Berlin, Heidelberg, 2010.
9.
IglesiasF.ZsebyT.FerreiraD. and ZimekA., MDCGen: multidimensional dataset generator for clustering, in: Journal of Classification36 (2019), 599–618.
10.
DemirtasH., A method for multivariate ordinal data generation given marginal distributions and correlations, in: Journal of Statistical Computation and Simulation76(11) (2006), 1017–1025.
11.
AdäI. and BertholdM.R., The New Iris Data: Modular Data Generators, in: KDD’10: Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2010, pp. 413–422.
12.
PearsonK., Notes on regression and inheritance in the case of two parents, in: Proceedings of the Royal Society of London58 (1895), 240–242.
13.
SpearmanC., The proof and measurement of association between two things, in: The American Journal of Psychology15(1) (1904), 441–471.
14.
KendallM.G., A new measure of rank correlation, in: Biometrika30(1–2) (1938), 81–93.
15.
XuW.HouY.HungY.S. and ZouY., Comparison of spearman’s rho and kendall’s tau in normal and contaminated normal models, in: Signal Processing93(1) (2013), 261–276.
16.
NelderJ.A. and MeadR., A simplex method for function minimization, in: The Computer Journal7(4) (1965), 308–313.
17.
DevroyeL., Non-Uniform Random Variate Generation, Springer, New York, 1986.
18.
HighamN.J., Computing a nearest symmetric positive semidefinite matrix, in: Linear Algebra and its Applications103 (1988), 103–118.
19.
RunklerT.A., Data Analytics: Models and Algorithms for Intelligent Data Analysis, Springer, Wiesbaden, 2012.
20.
ScruccaL.FopM.MurphyT.B. and RafteryA.E., Mclust 5: clustering, classification and density estimation using Gaussian finite mixture models, in: The R Journal8(1) (2016), 205–233.
21.
R Core Team, R: A language and environment for statistical computing. R Foundation for Statistical Computing, URL https://www.R-project.org/, 2019.
22.
ChenH.QuandtS.A.GrzywaczJ.G. and ArcuryT.A., A distribution-based multiple imputation method for handling bivariate pesticide data with values below the limit of detection, in: Environmental Health Perspectives119(3) (2011), 351–356.
23.
HarelO.PerkinsN. and SchistermanE.F., The use of multiple imputation for data subject to limits of detection, in: Sri Lankan Journal of Applied Statistics5(4) (2014), 227–246.
