SAMSI Home Page

Data Confidentiality
 A Working Group in National Defense and Homeland Security

 

Group Leader: Alan Karr (NISS)

 Announcements

Meeting Time: Tuesdays 9:15 - 10:45 am
Location: NISS building, 2nd floor conference room

Remote access for the meeting is available:

  • Via teleconference at 919-685-9338 not earlier than 9:00 am

Note: Until we can get the needed bandwidth, equipment, etc. for broadcasting meetings via a webcam, SAMSI will be employing the following process for remote participation in working groups. A webcam picture will be available for viewing during the meetings. However, for audio, participants will have to call into the conference line.

Description

The Data Confidentiality workgroup of the National Defense and Homeland Security SAMSI program focuses on several areas of confidentiality, including secure multiparty computations, synthetic data, disclosure risk and data utility. This group integrates naturally with an ongoing research at NISS, therefore having some momentum before the program started.

Meeting Activities

Date Details
May 2 2006 Communications:

Presentations:
April 25 2006 Communications:

Presentations:
April 18 2006 No meeting this day

Communications:
April 11 2006 No meeting this day
April 4 2006 No meeting this day
March 28 2006 Communications:

Articles:
March 21 2006 Communications:
March 14 2006 No meeting this day

Communications:
March 7 2006 Presentations:
Articles:
Feb 28 2006 Presentations:
Feb 21 2006 Presentations:
Feb 14 2006 Presentations:
Feb 7 2006 Presentations:
Jan 31 2006 Presentations:
Articles:
Jan 24 2006 Presentations:
Articles:
Jan 9 2006 Presentations:
Manuscripts:
Jan 3 2006 Communications:
Dec 2 2005 Communications:

Presentations:

Notes

  • Group decided that meetings next year will be on Tuesday at 9:00 am, starting on January 3
Nov 18 2005 Communications:

Presentations:

Articles:
Nov 11 2005

Communications:
Nov 4 2005

Articles:

Notes:

  • Software Demonstration
Oct 28 2005

Presentations:
Oct 21 2005

Presentations:
Oct 14 2005

Communications:

Articles:

Notes:

  • The propensity scores idea is discused further.
Oct 7 2005

Notes:

  • Propensity scores are discussed, with work between Mi-Ja and Anna. Anna has some microdata for which some of the techiniques of propensity scores are discussed
  • For propensity scores, there are problems with logistic regression models that include high degree polynomial terms. Bahjat suggests centering and possibly scaling
Sep 30 2005

Communications:
Sep 16 2005 Articles:

Notes:

  • There was some brain-storming on utility measures, including the idea of propensity scores presented by Mi-Ja, using some of the ideas by Rosenbaum.

 
Sep 9 2005

Notes:

  • The idea of propensity scores was presented. Mi-Ja was assigned to run some simulations and try the idea.
  • Some comments were made about how propensity scores fall under a more general framework of classification, where the idea is that the more difficult to classify real data from synthetic data the better the utility of the data.
Sep 2 2005

Articles:

Links of interest:


Group Members

Name Email Address Affiliation

Lisa Denogean

[email protected]

SAMSI and North Carolina State University
Joyee Ghosh [email protected] Duke University
Alan Karr [email protected] NISS
Myron J. Katzoff [email protected] Center for Disease Control
Saki Kinney [email protected] Duke University

James Lynch

[email protected]

University of South Carolina

Anna Oganian [email protected] NISS
Bahjat Qaqish

[email protected]

 University of North Carolina at Chapel Hill
Jerome Reiter [email protected] Duke University

Francisco Vera

[email protected]

NISS and SAMSI

Mi-Ja Woo [email protected] NISS

Lawrence H. Cox

 [email protected]

National Center for Health Statistics

Joe Fred Gonzalez, Jr.

 [email protected]

National Center for Health Statistics


If you are interested in participating in a working group, please send e-mail to [email protected] giving your full contact information and stating which working group(s). You will be sent instructions by e-mail about how to participate.

 

Send your comments about this site to Francisco Vera ([email protected]).

To subscribe or unsubscribe to the email list, contact Alan Karr ([email protected]).