HKUEMS :: Event Details

50th Anniv. Seminar Series on 'Some aspects of data analysis under ...' by Professor Bimal SINHA
posted by Department of Statistics and Actuarial Science for HKU and Public
Event Type: Public Lecture/Forum/Seminar/Workshop/Conference/Symposium
Event Nature: Science & Technology

Event Details

DEPARTMENT OF STATISTICS AND ACTUARIAL SCIENCE
THE UNIVERSITY OF HONG KONG

50th Anniversary Seminar Series

Professor Bimal SINHA
Department of Mathematics and Statistics
UMBC & Center for Statistical Research & Methodology (CSRM)
US Census Bureau
USA

will give a talk

entitled

SOME ASPECTS OF DATA ANALYSIS UNDER CONFIDENTIALITY PROTECTION

Abstract

A main focus of recent research in many government agencies (such as US Census Bureau) is the novel creation and valid analysis of perturbed data which are either magnitude or categorical in nature. US Census Bureau mandate calls for protection of confidential data, avoidance of identity disclosure, and valid analysis of eventually released data! Often some form of perturbation of original data is used through differential privacy approach, swapping, noise addition/multiplication, pramming (for qualitative variables), creating singly or multiply imputed synthetic data, and so on Research is needed to produce such variations of the original data, provide valid statistical analysis of the released data, and evaluate the level of confidentiality protection supplied by the underlying new data generation methods.

In these lectures I will touch upon two aspects of data perturbation: noise multiplication and synthetic data, and elaborate on how to create and sensibly analyze such perturbed data with some examples. Time permitting, I will also present some severe consequences of model misspecifications.

Date/Time22/08/2017 15:00-17:00
VenueRoom 301, Run Run Shaw Building, HKU
LanguageEnglish
Programmeclick to view
14:45 to 15:00 Refreshments will be served outside Room 301 Run Run Shaw Building
15:00 to 17:00 Seminar

Registration Instruction

Registration is not required.

Contact Information

For further information, please visit:
http://www.saasweb.hku.hk/seminar/seminar.php

Should you have any enquiries, please feel free to contact Ms. Irene Cheung by email at saas@hku.hk or by phone at 39173812 or by fax at 28589041.