BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Combining  statistical disclosure limitation methods to preserve r
 elationships  and data-specific  constraints in survey data. - Anna Ogania
 n (None / Other)
DTSTART:20161207T113000Z
DTEND:20161207T120000Z
UID:TALK69367@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:Applications of data swapping and noise are among the most wid
 ely used methods for Statistical Disclosure Limitation (SDL) by statistica
 l agencies for public-use non-interactive data release. The core ideas of 
 swapping and noise are conceptually easy to understand and are naturally s
 uited for masking purposes. We believe that they are worth revisiting with
  a special emphasis given to the utility aspects of these methods and to t
 he ways of combining the methods to increase their efficiency and reliabil
 ity. &nbsp\;Indeed\, many data collecting agencies use complex sample desi
 gns to increase the precision of their estimates and often allocate additi
 onal funds to obtain larger samples for particular groups in the populatio
 n. Thus\, it is particularly undesirable and counterproductive when SDL me
 thods applied to these data significantly change the magnitude of estimate
 s and/or their levels of precision. We will present and discuss two method
 s of disclosure limitation based on swapping and noise\, which can work to
 gether in synergy while protecting continuous and categorical variables. T
 he first method is a version of multiplicative noise that preserves means 
 and covariance together with some structural constraints in the data. The 
 second method is loosely based on swapping. It is designed with the goal o
 f preserving the relationships between strata-defining variables with othe
 r variables in the survey. We will show how these methods can be applied t
 ogether enhancing each other&rsquo\;s efficiency.       <br>
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR