1 / 8

Experiences of implementing Bifrost

Experiences of implementing Bifrost. Lars-Erik Almberg, Karin Andersson , Lei Sun Statistics Sweden Ottawa October 30, 2013. Introduction. Statistics Sweden has developed Bifrost as a standard tool for disclosure control Quality advantage Support advantage. BIFROST. HTML-report.

dragon
Télécharger la présentation

Experiences of implementing Bifrost

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Experiences of implementing Bifrost Lars-Erik Almberg, Karin Andersson, Lei Sun Statistics Sweden Ottawa October 30, 2013

  2. Introduction • Statistics Sweden has developed Bifrost as a standard tool for disclosure control • Quality advantage • Support advantage

  3. BIFROST HTML-report SAS macros Logfile Outputfile(s) τ-Argus Command sas2argus Metadata Data Xpress Design: • choice and parameterization of the methods Implementation: • configuration of and adaptation to various production systems (SAS, SQL, .NET, Excel, …) Adaptation layer

  4. Bifrost %sas2argus( Jobname= Table1, prefix for all text files InData = Microdataset, micro data Explanatory = Year Size, explanatory variables Response = Input, response variable SafetyRule = P(20,1), primary suppression Suppress = MOD(1,10), secondary suppression Out = Inter(1), specifies the output from τ-Argus RunArgus = 1, text files are created and τ-Argus is executed SAS = 2, import the output from τ-Argus toSAS work Debug = 1 information is written in the SAS-log )

  5. Consent • The Swedish Secrecy Act ensures confidentiality for collected data • In some cases Statistics Sweden asks businesses for their consent • Consent control is done in three steps: • First %sas2argus is run to check which cells are sensitive • Then a control is made to check if some of the unsafe cells could be set to safe • Finally %sas2argus is run once more to do secondary suppression

  6. Consent %consent_control( Jobname = Table1, prefix for all files Microdata = Microdataset, micro data Consentdata = Consentdataset, consent data Id = Controlnr, identification variable of the businesses SafetyRule = P, primary suppression rule SafetyVar = 20, primary suppression parameter Var1 = Year, explanatory variable Var2 = Size, explanatory variable Response = Input, response variable )

  7. Experiences • Bifrosthas so far been implemented in about 10 surveys • Paid from a central pot • Bifrost with consent control has been implemented in 8 surveys • Problems • Groups of businesses (holdings) • Tables without totals • Complex table structures

More Related