IDEAS home Printed from https://ideas.repec.org/p/arx/papers/2207.12602.html
   My bibliography  Save this paper

Differentially Private Estimation via Statistical Depth

Author

Listed:
  • Ryan Cumings-Menon

Abstract

Constructing a differentially private (DP) estimator requires deriving the maximum influence of an observation, which can be difficult in the absence of exogenous bounds on the input data or the estimator, especially in high dimensional settings. This paper shows that standard notions of statistical depth, i.e., halfspace depth and regression depth, are particularly advantageous in this regard, both in the sense that the maximum influence of a single observation is easy to analyze and that this value is typically low. This is used to motivate new approximate DP location and regression estimators using the maximizers of these two notions of statistical depth. A more computationally efficient variant of the approximate DP regression estimator is also provided. Also, to avoid requiring that users specify a priori bounds on the estimates and/or the observations, variants of these DP mechanisms are described that satisfy random differential privacy (RDP), which is a relaxation of differential privacy provided by Hall, Wasserman, and Rinaldo (2013). We also provide simulations of the two DP regression methods proposed here. The proposed estimators appear to perform favorably relative to the existing DP regression methods we consider in these simulations when either the sample size is at least 100-200 or the privacy-loss budget is sufficiently high.

Suggested Citation

  • Ryan Cumings-Menon, 2022. "Differentially Private Estimation via Statistical Depth," Papers 2207.12602, arXiv.org.
  • Handle: RePEc:arx:papers:2207.12602
    as

    Download full text from publisher

    File URL: http://arxiv.org/pdf/2207.12602
    File Function: Latest version
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Struyf, Anja & Rousseeuw, Peter J., 2000. "High-dimensional computation of the deepest location," Computational Statistics & Data Analysis, Elsevier, vol. 34(4), pages 415-426, October.
    2. Koenker, Roger, 2000. "Galton, Edgeworth, Frisch, and prospects for quantile regression in econometrics," Journal of Econometrics, Elsevier, vol. 95(2), pages 347-374, April.
    3. Jordan Awan & Aleksandra Slavković, 2021. "Structure and Sensitivity in Differential Privacy: Comparing K-Norm Mechanisms," Journal of the American Statistical Association, Taylor & Francis Journals, vol. 116(534), pages 935-954, April.
    4. Raj Chetty & John N. Friedman & Nathaniel Hendren & Maggie R. Jones & Sonya R. Porter, 2018. "The Opportunity Atlas: Mapping the Childhood Roots of Social Mobility," NBER Working Papers 25147, National Bureau of Economic Research, Inc.
    5. Massé, Jean-Claude, 2002. "Asymptotics for the Tukey Median," Journal of Multivariate Analysis, Elsevier, vol. 81(2), pages 286-300, May.
    6. Devroye, Luc, 1982. "Bounds for the uniform deviation of empirical measures," Journal of Multivariate Analysis, Elsevier, vol. 12(1), pages 72-79, March.
    7. Van Aelst, Stefan & Rousseeuw, Peter J. & Hubert, Mia & Struyf, Anja, 2002. "The Deepest Regression Method," Journal of Multivariate Analysis, Elsevier, vol. 81(1), pages 138-166, April.
    8. Wasserman, Larry & Zhou, Shuheng, 2010. "A Statistical Framework for Differential Privacy," Journal of the American Statistical Association, American Statistical Association, vol. 105(489), pages 375-389.
    9. Wei, Fan & Dudley, Richard M., 2012. "Two-sample Dvoretzky–Kiefer–Wolfowitz inequalities," Statistics & Probability Letters, Elsevier, vol. 82(3), pages 636-644.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Raj Chetty & John N. Friedman, 2019. "A Practical Method to Reduce Privacy Loss When Disclosing Statistics Based on Small Samples," AEA Papers and Proceedings, American Economic Association, vol. 109, pages 414-420, May.
    2. Xiaohui Liu & Shihua Luo & Yijun Zuo, 2020. "Some results on the computing of Tukey’s halfspace median," Statistical Papers, Springer, vol. 61(1), pages 303-316, February.
    3. Mia Hubert & Peter Rousseeuw & Pieter Segaert, 2015. "Multivariate functional outlier detection," Statistical Methods & Applications, Springer;Società Italiana di Statistica, vol. 24(2), pages 177-202, July.
    4. Hiau Joo Kee, 2005. "Glass Ceiling or Sticky Floor? Exploring the Australian Gender Pay Gap using Quantile Regression and Counterfactual Decomposition Methods," CEPR Discussion Papers 487, Centre for Economic Policy Research, Research School of Economics, Australian National University.
    5. Stephan Morgenthaler, 2007. "A survey of robust statistics," Statistical Methods & Applications, Springer;Società Italiana di Statistica, vol. 15(3), pages 271-293, February.
    6. Peter Leopold S. Bergman & Eric W. Chan & Adam Kapor, 2020. "Housing Search Frictions: Evidence from Detailed Search Data and a Field Experiment," CESifo Working Paper Series 8080, CESifo.
    7. John M. Abowd & Ian M. Schmutte & William Sexton & Lars Vilhuber, 2019. "Suboptimal Provision of Privacy and Statistical Accuracy When They are Public Goods," Papers 1906.09353, arXiv.org.
    8. John Gathergood & Fabian Gunzinger & Benedict Guttman-Kenney & Edika Quispe-Torreblanca & Neil Stewart, 2020. "Levelling Down and the COVID-19 Lockdowns: Uneven Regional Recovery in UK Consumer Spending," Papers 2012.09336, arXiv.org, revised Dec 2020.
    9. Pedro Telhado Pereira & Pedro Silva Martins, 2000. "Does education reduce wage inequality? Quantile regressions evidence from fifteen European countries," Nova SBE Working Paper Series wp379, Universidade Nova de Lisboa, Nova School of Business and Economics.
    10. Nathaniel Baum-Snow & Daniel Hartley & Kwan Ok Lee, 2019. "The Long-Run Effects of Neighborhood Change on Incumbent Families," Working Paper Series WP-2019-2, Federal Reserve Bank of Chicago.
    11. Magne Mogstad & Joseph P. Romano & Azeem Shaikh & Daniel Wilhelm, 2020. "Inference for Ranks with Applications to Mobility across Neighborhoods and Academic Achievement across Countries," NBER Working Papers 26883, National Bureau of Economic Research, Inc.
    12. Kristoffer Berg & Shafik Hebous, 2021. "Does A Wealth Tax Improve Equality of Opportunity?," CESifo Working Paper Series 9174, CESifo.
    13. Böhm, Michael Johannes & Gaudecker, Hans-Martin von & Schran, Felix, 2019. "Occupation Growth, Skill Prices, and Wage Inequality," IZA Discussion Papers 12647, Institute of Labor Economics (IZA).
    14. Ruth McAreavey & David L. Brown, 2019. "Comparative analysis of rural poverty and inequality in the UK and the US," Palgrave Communications, Palgrave Macmillan, vol. 5(1), pages 1-10, December.
    15. Debruyne, M. & Hubert, M. & Portnoy, S. & Vanden Branden, K., 2008. "Censored depth quantiles," Computational Statistics & Data Analysis, Elsevier, vol. 52(3), pages 1604-1614, January.
    16. Marcio Laurini, 2007. "A note on the use of quantile regression in beta convergence analysis," Economics Bulletin, AccessEcon, vol. 3(52), pages 1-8.
    17. Claire McKay Bowen & Fang Liu & Bingyue Su, 2021. "Differentially private data release via statistical election to partition sequentially," METRON, Springer;Sapienza Università di Roma, vol. 79(1), pages 1-31, April.
    18. Victoria Gregory & Julian Kozlowski & Hannah Rubinton, 2022. "The Impact of Racial Segregation on College Attainment in Spatial Equilibrium," Working Papers 2022-036, Federal Reserve Bank of St. Louis, revised 24 Jul 2023.
    19. Anastasia I. Koutsomanoli‐Filippaki & Emmanuel C. Mamatzakis, 2011. "Efficiency under quantile regression: What is the relationship with risk in the EU banking industry?," Review of Financial Economics, John Wiley & Sons, vol. 20(2), pages 84-95, May.
    20. Isaiah Andrews & Toru Kitagawa & Adam McCloskey, 2018. "Inference on winners," CeMMAP working papers CWP31/18, Centre for Microdata Methods and Practice, Institute for Fiscal Studies.

    More about this item

    NEP fields

    This paper has been announced in the following NEP Reports:

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:arx:papers:2207.12602. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: arXiv administrators (email available below). General contact details of provider: http://arxiv.org/ .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.