On the universal consistency of histograms anonymised by a randomised response technique

Martin Kroll

Statistics & Probability Letters, 2022, vol. 185, issue C

Abstract: The purpose of this short letter is to provide sufficient conditions for universal consistency (both pointwise and with respect to the L1-error) of density estimators under local differential privacy. As the object of study we consider the well-known histogram estimator anonymised by a randomised response technique in a multivariate setup. The universal consistency of this method has not been studied yet and, in contrast to existing work on other privacy mechanisms, we make the dependence on the privacy level in the conditions ensuring universal consistency explicit. Besides the stronger assumptions on the bandwidth parameter to obtain universal consistency, this approach also allows to record the effective reduction of the sample size present under local differential privacy.

Keywords: Multivariate density estimation; Local differential privacy; Strong universal consistency; Universal pointwise consistency; Randomised response (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0167715222000347
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:stapro:v:185:y:2022:i:c:s0167715222000347

Ordering information: This journal article can be ordered from
http://www.elsevier.com/wps/find/supportfaq.cws_home/regional
https://shop.elsevie ... _01_ooc_1&version=01

DOI: 10.1016/j.spl.2022.109420

Access Statistics for this article

Statistics & Probability Letters is currently edited by Somnath Datta and Hira L. Koul

More articles in Statistics & Probability Letters from Elsevier
Bibliographic data for series maintained by Catherine Liu ().