[Ml-stat-talks] Jerry Reiter

Kosuke Imai kimai at Princeton.EDU
Wed Mar 6 14:17:23 EST 2013


Hi,

  Jerry Reiter (Duke Statistics) is giving a talk at the political methodology seminar on this Friday about the confidentiality in public use data.

Best,
Kosuke

---------------------------------------------------------
Kosuke Imai               Office: Corwin Hall 036
Associate Professor       Phone: 609-258-6601
Department of Politics    Fax: 609-258-1110
Princeton University      Email: kimai at Princeton.Edu<mailto:kimai at Princeton.Edu>
Princeton, NJ 08544-1012  http://imai.princeton.edu
---------------------------------------------------------

Begin forwarded message:

From: Carlos Velasco <cvelasco at PRINCETON.EDU<mailto:cvelasco at PRINCETON.EDU>>
Subject: Methods Seminar this Week
Date: March 4, 2013 1:00:54 PM EST
To: <polmeth at Princeton.EDU<mailto:polmeth at Princeton.EDU>>
Reply-To: Carlos Velasco <cvelasco at PRINCETON.EDU<mailto:cvelasco at PRINCETON.EDU>>

Hi everyone,

This week in the methods seminar Jerry Reiter (Duke) will be giving a talk titled "Protecting Confidentiality in Public Use Data with Multiple Imputation". The abstract for the talk is:
"Many statistical agencies and survey organizations strive to release data to the public that (i) inform a wide range of analyses, (ii) protect confidentiality of data subjects, and (iii) are convenient to use. Meeting all three objectives is a difficult task. One approach is to release multiply-imputed, synthetic data sets in which sensitive or identifying values are replaced with values simulated from statistical models. This approach has been used, for example, by the Census Bureau to disseminate data from the Survey of Income and Program Participation, the Longitudinal Business Database, and the American Community Survey group quarters data. In this talk, I describe this novel use of multiple imputation. I review several recent applications and present the statistical theory needed to obtain valid inferences from the multiply-imputed data sets."
A copy of the paper for the talk and the the seminar's calendar for the rest of the semester can be found at http://q-aps.princeton.edu/book/political-methodology-research-seminar. If you would like to meet with Jerry Reiter next Friday please email me at cvelasco at princeton.edu<mailto:cvelasco at princeton.edu>. As always, we will be meeting on Friday at noon in Corwin 127. Lunch will be served.

Best,


Carlos




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cs.princeton.edu/pipermail/ml-stat-talks/attachments/20130306/61a5e1be/attachment.htm>


More information about the Ml-stat-talks mailing list