U.S. flag

An official website of the United States government

Skip Header


Constrained Noise for Masking Microdata Records

Written by:
RR90-04

Abstract

The objective of this report is to present two algorithms which transform data generated by a random number generator into data satisfying certain constraints on means and variance-covariance structure. One algorithm uses a linear transformation and translation to force data generated from a multivariate normal distribution to have a specific mean vector and variance-covariance matrix. The other algorithm uses a series of additions and subtractions to ensure that data generated from a uniform distribution has a certain mean and variance. Data sets such as these may be beneficial when used for introducing noise in order to mask microdata as a avoidance technique.

Related Information


Page Last Revised - October 28, 2021
Is this page helpful?
Thumbs Up Image Yes Thumbs Down Image No
NO THANKS
255 characters maximum 255 characters maximum reached
Thank you for your feedback.
Comments or suggestions?

Top

Back to Header