Efficiency and Sample Size Determination of Protected Data
RIS ID
130618
Abstract
This paper assesses the usefulness of a proposed multiplicative perturbation method by contrasting the statistical efficiency achieved in point hypothesis testing of simple proportions with that of the differentially private aggregated Laplace mechanism. This efficiency is evaluated by obtaining an analytical expression that determines the sample size required for protected data to retain a given significance level and power.
Publication Details
Wakefield, B. & Lin, Y. (2018). Efficiency and Sample Size Determination of Protected Data. Lecture Notes In Computer Scince, 11126 263-278.