pith. sign in

arxiv: 2606.12185 · v1 · pith:UHMY7EVWnew · submitted 2026-06-10 · 💰 econ.EM · math.ST· stat.TH

Pivotal and identification-robust nonparametric inference in linear IV models

Bertille Antoine , Pascal Lavergne This is my paper

Pith reviewed 2026-06-27 07:42 UTC · model grok-4.3

classification 💰 econ.EM math.STstat.TH
keywords linear IV modelsweak identificationheteroskedasticitynonparametric inferencepivotal testsidentification-robust inferencesubvector inferencespecification test
0
0 comments X

The pith

A modified test statistic for linear IV models is asymptotically pivotal under weak identification and unknown heteroskedasticity.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper develops inference procedures for linear instrumental variable models that remain valid whether identification is strong or weak and that allow for heteroskedasticity of unknown form while estimating the first stage nonparametrically. The central step is a direct modification of an existing test statistic so that it accounts for heteroskedasticity, which makes the statistic asymptotically pivotal and removes the need for simulation or case-by-case critical values. The resulting procedures cover inference on endogenous regressors, subvector tests that do not require knowledge of identification strength for nuisance parameters, and a specification test; all are shown to be conservative yet powerful. Practitioners gain methods that are both nonparametric in the first stage and easy to compute.

Core claim

By modifying an existing test statistic to directly account for heteroskedasticity of unknown form, the resulting statistic becomes asymptotically pivotal, which enables straightforward inference on parameters of endogenous variables, subvector inference without knowledge of identification strength, and a pure specification test in linear IV models that are nonparametric in the first stage.

What carries the argument

The modified test statistic that directly accounts for heteroskedasticity of unknown form and becomes asymptotically pivotal for the parameters of interest.

If this is right

  • Inference on parameters of endogenous explanatory variables can use standard asymptotic critical values without further adjustment.
  • Subvector inference remains valid without requiring knowledge of identification strength for the remaining parameters.
  • A pure specification test can be performed that stays conservative under weak identification.
  • The procedures remain computationally straightforward and can be applied with nonparametric first-stage estimation.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • These pivotal statistics might be adapted to other semiparametric IV settings where heteroskedasticity interacts with identification.
  • Applied researchers could apply the methods to recheck conclusions in areas prone to weak instruments, such as demand estimation or policy evaluation.
  • Extensions could examine whether similar modifications improve robustness in panel or time-series IV contexts.

Load-bearing premise

The linear IV model must be correctly specified with valid instruments and the nonparametric first-stage estimator must satisfy the regularity conditions needed for asymptotic pivotality.

What would settle it

A correctly specified data-generating process with heteroskedasticity and weak instruments in which the new statistic's finite-sample rejection rate under the null deviates substantially from the nominal level would falsify the asymptotic pivotality claim.

Figures

Figures reproduced from arXiv: 2606.12185 by Bertille Antoine, Pascal Lavergne.

Figure 1
Figure 1. Figure 1: Power curves for polynomial model with c = 3 and n = 201 (top left), with stronger identification c = 7 (top right), linear model with c = 3 and n = 201 (bottom left), polynomial group model with c = 3 and n = 401 (bottom right). equal to the number of instruments, so one obtains simple conservative critical values. We also compute the jackknife T-specification test proposed by Chao et al. (2014), hereafte… view at source ↗
Figure 2
Figure 2. Figure 2: Power curves for polynomial model with c = 3 and n = 201 (top left), with stronger identification c = 7 (top right), linear model with c = 3 and n = 201 (bottom left), polynomial group model with c = 3 and n = 401 (bottom right). group model. Throughout the different designs, the small sample properties of both JackT and JCUE are affected by the number of instruments. By constrast, our specification test c… view at source ↗
Figure 3
Figure 3. Figure 3: 95% confidence regions for the (nonlinear) effect of the population collapse obtained with HICM (left) and S (right). Specification Linear S2 S3 β1 [-1.23, -0.80] b1 [-1.24, -0.38] c1 [-2.50, 0.15] b2 [-5.20, -0.55] c2 [-7.20, -0.05] c3 [-17.40, 4.40] [PITH_FULL_IMAGE:figures/full_fig_p025_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: 95% confidence regions for the coefficients of the population collapse obtained [PITH_FULL_IMAGE:figures/full_fig_p038_4.png] view at source ↗
read the original abstract

We develop new inference procedures for a linear IV model that are robust to identification strength and heteroskedasticity of unknown form, and nonparametric with respect to the first-stage equation. Our first test is tailored for inference on parameters of endogenous explanatory variables. Our new statistic modifies that of Antoine and Lavergne (2003) to directly account for heteroskedasticity of unknown form. As a result, it is asymptotically pivotal, so that inference is greatly facilitated in practice. We also develop (i) an identification-robust subvector inference procedure that does not rely on the knowledge of identification strength for the remaining parameters, and (ii) a pure specification test. In both cases, the tests are conservative but powerful. We show that our procedures are computationally friendly and competitive with existing ones in simulations and an application.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 3 minor

Summary. The manuscript develops inference procedures for linear IV models that are robust to identification strength and heteroskedasticity of unknown form, while treating the first-stage equation nonparametrically. The core contribution modifies the Antoine and Lavergne (2003) statistic to incorporate heteroskedasticity directly, yielding an asymptotically pivotal test for parameters on endogenous regressors. Additional procedures include an identification-robust subvector test (conservative but powerful, without requiring knowledge of identification strength for other parameters) and a pure specification test. The methods are claimed to be computationally friendly and to perform competitively in simulations and an empirical application.

Significance. If the asymptotic pivotality and robustness results hold under the stated regularity conditions, the work supplies practical tools for IV inference that avoid nuisance-parameter estimation and remain valid under weak identification. The nonparametric first-stage broadens applicability beyond parametric assumptions. Credit is due for emphasizing computational feasibility alongside simulation and application evidence.

minor comments (3)
  1. [Abstract] Abstract: the statement that the modified statistic 'is asymptotically pivotal, so that inference is greatly facilitated in practice' would benefit from a brief parenthetical reference to the specific limiting distribution (e.g., standard normal or chi-squared) to make the practical advantage immediately clear.
  2. The regularity conditions on the nonparametric first-stage estimator are invoked for both strong and weak identification; a short table or paragraph summarizing the key rate and smoothness requirements would aid readers who wish to verify applicability to their data.
  3. The subvector and specification-test procedures are described as conservative; explicit finite-sample or local-power comparisons with existing conservative methods (e.g., Anderson-Rubin or Kleibergen) would strengthen the claim of competitiveness.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the constructive and positive report, which recognizes the paper's contributions to asymptotically pivotal and identification-robust inference in linear IV models under nonparametric first-stage and unknown heteroskedasticity. We appreciate the recommendation for minor revision and will incorporate any editorial suggestions in the revised version.

Circularity Check

0 steps flagged

Minor self-citation to 2003 base statistic; central heteroskedasticity modification is independent

full rationale

The paper modifies the Antoine-Lavergne (2003) statistic to incorporate unknown heteroskedasticity, yielding an asymptotically pivotal test under stated regularity conditions. This modification is presented as a new derivation rather than a direct reuse or fit of prior results. The 2003 citation serves as the starting point for extension but is not load-bearing for the pivotal property itself, which follows from the new accounting for heteroskedasticity. No self-definitional loops, fitted inputs renamed as predictions, or uniqueness theorems imported from the authors' own prior work appear in the abstract or claimed chain. The derivation remains self-contained against external benchmarks once the nonparametric first-stage regularity conditions are granted.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Based on abstract only; no explicit free parameters, invented entities, or ad-hoc axioms listed. Relies on standard asymptotic regularity conditions and the linear IV model setup.

axioms (2)
  • standard math Standard regularity conditions for asymptotic theory in nonparametric estimation and weak identification
    Required for the pivotal property and robustness claims to hold.
  • domain assumption Linear IV model is correctly specified with valid instruments
    Foundation for all inference procedures described.

pith-pipeline@v0.9.1-grok · 5663 in / 1220 out tokens · 26504 ms · 2026-06-27T07:42:09.115336+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

78 extracted references · 41 canonical work pages

  1. [1]

    Andrews, Donald W. K. and Moreira, Marcelo J. and Stock, James H. , journal =. Optimal Two-Sided Invariant Similar Tests for Instrumental Variables Regression , year =. doi:10.1111/j.1468-0262.2006.00680.x , keywords =

    work page doi:10.1111/j.1468-0262.2006.00680.x 2006

  2. [2]

    Anderson, T. W. and Rubin, Herman , journal =. The Asymptotic Properties of Estimates of the Parameters of a Single Equation in a Complete System of Stochastic Equations , year =

  3. [3]

    Conditional Inference With a Functional Nuisance Parameter , year =

    Andrews, Isaiah and Mikusheva, Anna , journal =. Conditional Inference With a Functional Nuisance Parameter , year =. doi:10.3982/ECTA12868 , file =

    work page doi:10.3982/ecta12868

  4. [4]

    A Geometric Approach to Nonlinear Econometric Models , year =

    Andrews, Isaiah and Mikusheva, Anna , journal =. A Geometric Approach to Nonlinear Econometric Models , year =. doi:10.3982/ECTA12030 , file =

    work page doi:10.3982/ecta12030

  5. [5]

    Andrews, Donald W. K. and Guggenberger, Patrik , journal =. Identification and Singularity-Robust Inference for Moment Condition Models , year =. doi:10.3982/QE1219 , file =

    work page doi:10.3982/qe1219

  6. [6]

    Andrews, Donald W. K. , journal =. Nonparametric Kernel Estimation for Semiparametric Models , year =. doi:10.1017/S0266466600009427 , file =

    work page doi:10.1017/s0266466600009427

  7. [7]

    Andrews, Donald W. K. and Marmer, Vadim and Yu, Zhengfei , journal =. On Optimal Inference in the Linear IV Model , year =. doi:10.3982/QE1082 , file =

    work page doi:10.3982/qe1082

  8. [8]

    Andrews, Donald W. K. and Cheng, Xu , journal =. Estimation and Inference With Weak, Semi-Strong, and Strong Identification , year =. doi:10.3982/ECTA9456 , keywords =

    work page doi:10.3982/ecta9456

  9. [9]

    Conditional Linear Combination Tests for Weakly Identified Models , year =

    Andrews, Isaiah , journal =. Conditional Linear Combination Tests for Weakly Identified Models , year =. doi:10.3982/ECTA12407 , keywords =

    work page doi:10.3982/ecta12407

  10. [10]

    Andrews, Donald W. K. , booktitle =. Empirical Process Methods in Econometrics , year =. doi:10.1016/S1573-4412(05)80006-6 , issn =

    work page doi:10.1016/s1573-4412(05)80006-6

  11. [11]

    Antoine and P

    B. Antoine and P. Lavergne , journal =. Conditional Moment Models under Semi-Strong Identification , year =

  12. [12]

    Anderson, T. W. and Rubin, Herman , journal =. Estimation of the Parameters of a Single Equation in a Complete System of Stochastic Equations , year =

  13. [13]

    D. W. K. Andrews and J. H. Stock , booktitle =. Inference with Weak Instruments , year =

  14. [14]

    Bierens , journal =

    H.J. Bierens , journal =. Consistent Model Specification Tests , year =

  15. [15]

    Bierens , journal =

    Herman J. Bierens , journal =. A Consistent Conditional Moment Test of Functional Form , year =

  16. [16]

    Bierens and Werner Ploberger , journal =

    Herman J. Bierens and Werner Ploberger , journal =. Asymptotic Theory of Integrated Conditional Moment Tests , year =

  17. [17]

    Burbage and Lonnie Magee and A

    John B. Burbage and Lonnie Magee and A. Leslie Robb , journal =. Alternative Transformations to Handle Extreme Values of the Dependent Variable , year =

  18. [18]

    and Farrell, Max H

    Cattaneo, Matias D. and Farrell, Max H. , journal =. Optimal Convergence Rates, Bahadur Representation, and Asymptotic Normality of Partitioning Estimators , year =. doi:10.1016/j.jeconom.2013.02.002 , file =

    work page doi:10.1016/j.jeconom.2013.02.002 2013

  19. [19]

    The Reduced Form: A Simple Approach to Inference with Weak Instruments , year =

    Victor Chernozhukov and Christian Hansen , journal =. The Reduced Form: A Simple Approach to Inference with Weak Instruments , year =. doi:10.1016/j.econlet.2007.11.012 , keywords =

    work page doi:10.1016/j.econlet.2007.11.012 2007

  20. [20]

    Admissible Invariant Similar Tests for Instrumental Variables Regression , year =

    Chernozhukov, Victor and Hansen, Christian and Jansson, Michael , journal =. Admissible Invariant Similar Tests for Instrumental Variables Regression , year =. doi:10.1017/S0266466608090312 , file =

    work page doi:10.1017/s0266466608090312

  21. [21]

    and Venter, J

    de Wet, T. and Venter, J. H. , journal =. Asymptotic Distributions for Quadratic Forms with Applications to Tests of Fit , year =. doi:10.1214/aos/1176342378 , fjournal =

    work page doi:10.1214/aos/1176342378

  22. [22]

    and Snell, Andy , journal =

    Dieterle, Steven G. and Snell, Andy , journal =. A Simple Diagnostic to Investigate Instrument Validity and Heterogeneous Effects When Using a Single Instrument , year =. doi:10.1016/j.labeco.2016.08. , fjournal =

    work page doi:10.1016/j.labeco.2016.08 2016

  23. [23]

    Dreier and S

    I. Dreier and S. Kotz , journal =. A Note on the Characteristic Function of the T-Distribution , year =. doi:10.1016/S0167-7152(02)00032-9 , fjournal =

    work page doi:10.1016/s0167-7152(02)00032-9

  24. [24]

    Dufour , journal =

    J.-M. Dufour , journal =. Identification, Weak Instruments, and Statistical Inference in Econometrics , year =

  25. [25]

    Further Results on Projection-Based Inference in IV Regressions with Weak, Collinear or Missing Instruments , year =

    Dufour, Jean-Marie and Taamouti, Mohamed , journal =. Further Results on Projection-Based Inference in IV Regressions with Weak, Collinear or Missing Instruments , year =. doi:10.1016/j.jeconom.2006.06.008 , file =

    work page doi:10.1016/j.jeconom.2006.06.008 2006

  26. [26]

    Romano and Michael Wolf

    Dufour, Jean-Marie and Taamouti, Mohamed , journal =. Projection-. 2005 , issn =. doi:10.1111/j.1468-0262.2005.00618.x , file =

    work page doi:10.1111/j.1468-0262.2005.00618.x 2005

  27. [27]

    Dufour , journal =

    J.-M. Dufour , journal =. Some Impossibility Theorems in Econometrics with Applications to Structural and Dynamic Models. , year =

  28. [28]

    Han and P.C.B

    C. Han and P.C.B. Phillips , journal =. GMM with Many Moment Conditions , year =

  29. [29]

    and Newey, Whitney K

    Hausman, Jerry A. and Newey, Whitney K. and Woutersen, Tiemen and Chao, John C. and Swanson, Norman R. , journal =. Instrumental Variable Estimation with Heteroskedasticity and Many Instruments , year =. doi:10.3982/QE89 , file =

    work page doi:10.3982/qe89

  30. [30]

    Hahn and J

    J. Hahn and J. Hausman , journal =. Weak Instruments: Diagnosis and Cures in Empirical Economics , year =

  31. [31]

    Hansen and J.A

    C. Hansen and J.A. Hausman and W.K. Newey , journal =. Estimation With Many Instrumental Variables , year =

  32. [32]

    and Kotz, S

    Johnson, N.L. and Kotz, S. and Balakrishnan, N. , publisher =. Continuous Univariate Distributions , year =

  33. [33]

    Jun, S. J. and Pinkse, J. , journal =. Testing Under Weak Identification with Conditional Moment Restrictions , year =. doi:10.1017/s0266466612000138 , fjournal =

    work page doi:10.1017/s0266466612000138

  34. [34]

    Uniformity and the Delta Method , year =

    Kasy, Maximilian , journal =. Uniformity and the Delta Method , year =. doi:10.1515/jem-2018-0001 , file =

    work page doi:10.1515/jem-2018-0001 2018

  35. [35]

    Kleibergen , journal =

    F. Kleibergen , journal =. Testing Parameters in. 2005 , pages =

    2005

  36. [36]

    Generalizing Weak Instrument Robust IV Statistics Towards Multiple Parameters, Unrestricted Covariance Matrices and Identification Statistics , year =

    Kleibergen, Frank , journal =. Generalizing Weak Instrument Robust IV Statistics Towards Multiple Parameters, Unrestricted Covariance Matrices and Identification Statistics , year =. doi:10.1016/j.jeconom.2006.06.010 , file =

    work page doi:10.1016/j.jeconom.2006.06.010 2006

  37. [37]

    Pivotal Statistics for Testing Structural Parameters in Instrumental Variables Regression , year =

    Kleibergen, Frank , journal =. Pivotal Statistics for Testing Structural Parameters in Instrumental Variables Regression , year =. doi:10.1111/1468-0262.00353/full , file =

    work page doi:10.1111/1468-0262.00353/full

  38. [38]

    2008 , Series =

    Introduction to Empirical Processes and Semiparametric Inference , Author =. 2008 , Series =

    2008

  39. [39]

    and Romano, J.P

    Lehmann, E.L. and Romano, J.P. , publisher =. Testing Statistical Hypotheses , year =

  40. [40]

    Smooth Minimum Distance Estimation and Testing with Conditional Estimating Equations: Uniform in Bandwidth Theory , year =

    Lavergne, Pascal and Patilea, Valentin , journal =. Smooth Minimum Distance Estimation and Testing with Conditional Estimating Equations: Uniform in Bandwidth Theory , year =. doi:10.1016/j.jeconom.2013.05.006 , type =

    work page doi:10.1016/j.jeconom.2013.05.006 2013

  41. [41]

    and Alix-Garcia, Jennifer , journal =

    Sellars, Emily A. and Alix-Garcia, Jennifer , journal =. Labor scarcity, land tenure, and historical legacy: Evidence from Mexico , year =. doi:10.1016/j.jdeveco.2018.07.014 , file =

    work page doi:10.1016/j.jdeveco.2018.07.014 2018

  42. [42]

    Robust Confidence Sets in the Presence of Weak Instruments , year =

    Anna Mikusheva , journal =. Robust Confidence Sets in the Presence of Weak Instruments , year =. doi:10.1016/j.jeconom.2009.12.003 , keywords =

    work page doi:10.1016/j.jeconom.2009.12.003 2009

  43. [43]

    M. J. Moreira , journal =. A Conditional Likelihood Ratio Test for Structural Models , year =

  44. [44]

    , journal =

    Moreira, Humberto and Moreira, Marcelo J. , journal =. Optimal Two-Sided Tests for Instrumental Variables Regression with Heteroskedastic and Autocorrelated Errors , year =. doi:10.1016/j.jeconom.2019.04.038 , keywords =

    work page doi:10.1016/j.jeconom.2019.04.038 2019

  45. [45]

    2017 , month = apr, adsnote =

    Working paper, FPG , title =. 2017 , month = apr, adsnote =. arXiv , doi =:1705.00231 , keywords =

    arXiv 2017

  46. [46]

    Nonparametric Covariance Model , year =

    Jianxin Yin and Zhi Geng and Runze Li and Hansheng Wang , journal =. Nonparametric Covariance Model , year =

  47. [47]

    Newey and F

    W.K. Newey and F. Windmeijer , journal =. Generalized Method of Moments With Many Weak Moment Conditions , year =. doi:10.3982/ECTA6224 , file =

    work page doi:10.3982/ecta6224

  48. [48]

    Srivastava

    Rice, John , journal =. Bandwidth Choice for Nonparametric Regression , year =. doi:10.1214/aos/1176346788 , fjournal =

    work page doi:10.1214/aos/1176346788

  49. [49]

    Adaptive nonparametric confidence sets , volume =

    Robins, James and van der Vaart, Aad , journal =. Adaptive Nonparametric Confidence Sets , year =. doi:10.1214/009053605000000877 , file =

    work page doi:10.1214/009053605000000877

  50. [50]

    Limit Theorems for Polylinear Forms , year =

    V.I Rotar' , journal =. Limit Theorems for Polylinear Forms , year =. doi:10.1016/0047-259X(79)90055-1 , keywords =

    work page doi:10.1016/0047-259x(79)90055-1

  51. [51]

    Nonparametric Estimation of Residual Variance Revisited , year =

    Burkhardt Seifert and Theo Gasser and Andreas Wolf , journal =. Nonparametric Estimation of Residual Variance Revisited , year =

  52. [52]

    Staiger and J

    D. Staiger and J. H. Stock , journal =. Instrumental Variables Regression with Weak Instruments , year =

  53. [53]

    and White, Halbert , journal =

    Stinchcombe, Maxwell B. and White, Halbert , journal =. Consistent Specification Testing With Nuisance Parameters Present Only Under the Alternative , year =

  54. [54]

    J. H. Stock and J. H. Wright , journal =. 2000 , pages =

    2000

  55. [55]

    J. H. Stock and J. H. Wright and M. Yogo , journal =. A Survey of Weak Instruments and Weak Identification in Generalized Method of Moments , year =

  56. [56]

    van der Vaart, A. W. and Wellner, Jon A , publisher =. Weak Convergence and Empirical Processes: with Applications to Statistics , year =

  57. [57]

    Bracketing Smooth Functions , year =

    van der Vaart, Aad , journal =. Bracketing Smooth Functions , year =

  58. [58]

    2017 , author =

    Generalized Additive Models: An Introduction with R , publisher =. 2017 , author =

    2017

  59. [59]

    A simple and robust estimator for linear regression models with strictly exogenous instruments , year =

    Escanciano, Juan Carlos , journal =. A simple and robust estimator for linear regression models with strictly exogenous instruments , year =. doi:10.1111/ectj.12087 , file =

    work page doi:10.1111/ectj.12087

  60. [60]

    Charles F

    Dominguez, Manuel A. and Lobato, Ignacio N. , journal =. Consistent Estimation of Models Defined by Conditional Moment Restrictions , year =. doi:10.1111/j.1468-0262.2004.00545.x , file =

    work page doi:10.1111/j.1468-0262.2004.00545.x 2004

  61. [61]

    2021 , month = jan, abstract =

    Chen, Xiaohong and Lee, Sokbae and Seo, Myung Hwan , title =. 2021 , month = jan, abstract =. arxiv , doi =:2008.11140 , file =

    arXiv 2021

  62. [62]

    Economet

    Montiel Olea, Jos. Economet. Theor. , title =. 2020 , pages =

    2020

  63. [63]

    Mathematical Statistics , year =

    Jun Shao , publisher =. Mathematical Statistics , year =

  64. [64]

    Identification-robust nonparametric inference in a linear

    Antoine, Bertille and Lavergne, Pascal , journal =. Identification-robust nonparametric inference in a linear. 2023 , issn =. doi:10.1016/j.jeconom.2022.01.011 , file =

    work page doi:10.1016/j.jeconom.2022.01.011 2023

  65. [65]

    On the instrument functional form with a binary endogenous explanatory variable , year =

    Xu, Ruonan , journal =. On the instrument functional form with a binary endogenous explanatory variable , year =. doi:10.1016/j.econlet.2021.109993 , file =

    work page doi:10.1016/j.econlet.2021.109993 2021

  66. [66]

    A Consistent

    Jiang, Feiyu and Tsyawo, Emmanuel Selorm , journal =. A Consistent. 2024 , abstract =. arxiv , doi =:2208.13370 , file =

    arXiv 2024

  67. [67]

    Feasible IV regression without excluded instruments , year =

    Tsyawo, Emmanuel Selorm , journal =. Feasible IV regression without excluded instruments , year =. doi:10.1093/ectj/utac032 , file =

    work page doi:10.1093/ectj/utac032

  68. [68]

    Instrumental variable estimation with first-stage heterogeneity , year =

    Abadie, Alberto and Gu, Jiaying and Shen, Shu , journal =. Instrumental variable estimation with first-stage heterogeneity , year =. doi:10.1016/j.jeconom.2023.02.005 , file =

    work page doi:10.1016/j.jeconom.2023.02.005 2023

  69. [69]

    J. D. Angrist and G. W. Imbens and A. B. Krueger , journal =. Jackknife instrumental variables estimation , year =

  70. [70]

    Guggenberger and F

    P. Guggenberger and F. Kleibergen and S. Mavroeidis , journal =. A more powerful subvector Anderson-Rubin test in linear variables regression , year =

  71. [71]

    Tanaka and C

    T. Tanaka and C. Camerer and Q. Ngyuen , journal =. Risk and Time Preferences: Linking Experimental and Household Survey Data from Vietnam , year =

  72. [72]

    Antoine and D

    B. Antoine and D. Frazier and E. Renault , journal =. Coordinated Testing for Identification Failure and Correct Model Specification , year =

  73. [73]

    Kleibergen and Z

    F. Kleibergen and Z. Zhan , journal =. Double robust inference for continuous updating. 2025 , pages =

    2025

  74. [74]

    Inference for Parameters Identified by Conditional Moment Restrictions Using a Generalized Bierens Maximum Statistic , year =

    Chen, Xiaohong and Lee, Sokbae and Seo, Myung Hwan and Song, Myunghyun , journal =. Inference for Parameters Identified by Conditional Moment Restrictions Using a Generalized Bierens Maximum Statistic , year =. doi:10.1162/rest_a_01550 , fjournal =

    work page doi:10.1162/rest_a_01550

  75. [75]

    Inference with Many Weak Instruments , year =

    Mikusheva, Anna and Sun, Liyang , journal =. Inference with Many Weak Instruments , year =. doi:10.1093/restud/rdab097 , file =

    work page doi:10.1093/restud/rdab097

  76. [76]

    and Hausman, Jerry A

    Chao, John C. and Hausman, Jerry A. and Newey, Whitney K. and Swanson, Norman R. and Woutersen, Tiemen , journal =. Testing overidentifying restrictions with many instruments and heteroskedasticity , year =. doi:10.1016/j.jeconom.2013.08.003 , file =

    work page doi:10.1016/j.jeconom.2013.08.003 2013

  77. [77]

    and Sun, Liyang , journal =

    Andrews, Isaiah and Stock, James H. and Sun, Liyang , journal =. Weak. 2019 , number =. doi:10.1146/annurev-economics-080218-025643 , file =

    work page doi:10.1146/annurev-economics-080218-025643 2019

  78. [78]

    A uniformly valid test for instrument exogeneity , year =

    Dovonon, Prosper and Gospodinov, Nikolay , journal =. A uniformly valid test for instrument exogeneity , year =. doi:10.1016/j.jeconom.2026.106231 , file =

    work page doi:10.1016/j.jeconom.2026.106231 2026