Skip Header

Simulation Study of the Effectiveness of Masking Microdata with Mixtures of Multivariate Normal Distributions

Sam Hawala
Component ID: #ti1764678403

Abstract

Continuous variables in microdata can be masked for protection from disclosure through the use of an additive noise. I consider adding noise that is distributed according to a mixture of normal distributions. There are several parameters involved in constructing the additive noise. The study’s purpose is to lay down as a guide a recipe for the choices of these parameters. The proportion of re-identifiable records through the use of Winkler’s matching software measures effectiveness of the masking method. Results depend heavily on the matching software used.

Related Information


X
  Is this page helpful?
Thumbs Up Image Yes    Thumbs Down Image No
X
Comments or suggestions?
No, thanks
255 characters remaining
X
Thank you for your feedback.
Comments or suggestions?
Back to Header