Statistics Colloquium : Dr. Bimal Sinha

Title:

Data Analysis Under Differential Privacy

Abstract:

Statistical disclosure control (SDC) methodology, often used by statistical agencies, aims to modify a sensitive dataset prior to its release to the public. The goal is not to disclose confidential information gathered from respondents while maintaining data utility. In this talk I will discuss the notion of Differential Privacy as applied to SDC method, and describe SDC method based on noise addition for data perturbation which is differentially private. I will end with applications  of multiple imputation methods for data analysis.