From Opasnet
Jump to: navigation, search


How to generate an artificial data from a real data in such a way that

  • individuals are no longer identifiable from the data,
  • the marginal distributions are close to those of the original data,
  • the correlation structure is close to that of the original data?


Rcode ReplicaX by Juha Karvanen does exactly that.

⇤--arg1500: . Table data cannot be pasted to the code for some reason. --Jouni (talk) 11:33, 18 November 2018 (UTC) (type: truth; paradigms: science: attack)

Give the variable types in format c('discrete','continuous'):

Give the k parameter (default recommended):

Give the osigma parameter (default recommended):

Do you want to show the replica data on screen?:

+ Show code

Show example data and analysis (Making a new run takes ca. 4 minutes)

+ Show code



Replica R function by Juha Karvanen does exactly that.

+ Show code

See also