statshorts

A conference paper I co-authored (together with Johannes Gussenbauer, Julien Jamme, Edwin de Jonge, and Peter-Paul de Wolf) titled ' Spatial SDC experiments and evaluations with multiple countries comparison ' is available online . The paper was prepared for the 2023 UNECE Expert Meeting on Statistical Data Confidentiality and presented in the session Challenges in publishing safe tables and maps . Quoting the abstract: "This study utilizes a standardized 'census-like' dataset that is structured uniformly across all participating countries to assess disclosure risk based on grid data . We begin by evaluating and comparing the risk using this approach. Next, we apply spatial SDC methods from the R package sdcSpatial, including kernel density smoothing and quad tree aggregation. We re-evaluate the disclosure risk using these methods and analyze the resulting utility loss . Our analysis will be conducted across multiple countries , allowing for a comprehensive compar...

In a previous blog post I have considered the case of a regression problem, where the distance between points is a covariate and one class of points is randomly displaced for privacy protection. This setup was directly derived from the Demographic and Health Surveys (DHS) program, which uses such a displacement procedure (Burgert et al., 2013). I have further stated that the standard measurement error model may not be admissible, since the error is biased . This is in direct contradiction to Warren et al. (2016), who do not only state that the standard measurement error model may be used, but also claim to show that its assumptions (including unbiased error) hold: "We also show that the observed distance covariate follows the classical measurement error model form [...] with $\mathrm{E}(u_i) = 0, \mathrm{Var}(u_i) = \sigma^2_u$". (Warren et al., 2016) Their analysis is a dedicated companion piece to the DHS data publication and reappears in the official guidelines docume...