EconBase
← All papers

Differentially Private Two-Stage Gradient Descent for Instrumental Variable Regression

Haodong Liang, Yanhao Jin, Krishnakumar Balasubramanian, Lifeng Lai

arXiv 26 Sep 2025 · Statistics — Machine Learning

arXiv:2509.22794 · PDF · DOI · OpenAlex · Extracted main text

Abstract

We study instrumental variable regression (IVaR) under differential privacy constraints. Classical IVaR methods (like two-stage least squares regression) rely on solving moment equations that directly use sensitive covariates and instruments, creating significant risks of privacy leakage and posing challenges in designing algorithms that are both statistically efficient and differentially private. We propose a noisy two-state gradient descent algorithm that ensures $\rho$-zero-concentrated differential privacy by injecting carefully calibrated noise into the gradient updates. Our analysis establishes finite-sample convergence rates for the proposed method, showing that the algorithm achieves consistency while preserving privacy. In particular, we derive precise bounds quantifying the trade-off among privacy parameters, sample size, and iteration-complexity. To the best of our knowledge, this is the first work to provide both privacy guarantees and provable convergence rates for instrumental variable regression in linear models. We further validate our theoretical findings with experiments on both synthetic and real datasets, demonstrating that our method offers practical accuracy-privacy trade-offs.

Citation extraction

42
references
56
in-text mentions
42
distinct cited
2
self-citations
7,143
main-text words

appendix boundary found by appendix_command · 31% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1Angrist, Joshua D. and Krueger, Alan B (2001) Instrumental Variables and the Search for Identification: From Supply and Demand to Natural Experiments0.73732100%
2Bun, Mark and Steinke, Thomas (2016) Concentrated differential privacy: Simplifications, extensions, and lower bounds0.73732100%
3Dwork, Cynthia and McSherry, Frank and Nissim, Kobbi and Smith, Adam (2006) Calibrating noise to sensitivity in private data analysis0.64422100%
4Bassily, Raef and Smith, Adam and Thakurta, Abhradeep (2014) Private empirical risk minimization: Efficient algorithms and tight error bounds0.58531100%
5Tsfadia, Eliad and Cohen, Edith and Kaplan, Haim and Mansour, Yishay… (2022) Friendlycore: Practical differentially private aggregation0.5114225%
6Vershynin, Roman (2018) High-Dimensional Probability: An Introduction with Applications in Data Science0.5112250%
7Abadi, Martin and Chu, Andy and Goodfellow, Ian and McMahan, H Brend… (2016) Deep learning with differential privacy0.51121100%
8Sheffet, Or (2017) Differentially private ordinary least squares0.51121100%
9Wang, Yu-Xiang and Balle, Borja and Kasiviswanathan, Shiva Prasad (2019) Subsampled rényi differential privacy and analytical moments accountant0.51121100%
10Westoff, Charles F. and Parke, Robert (1972) Demographic and social aspects of population growth0.40511100%

Showing the top 10 of 42 scored citations.