pith. sign in

arxiv: 2604.07718 · v1 · submitted 2026-04-09 · 💰 econ.EM · math.ST· stat.TH

Identification in (Endogenously) Nonlinear SVARs Is Easier Than You Think

James A. Duffy , Sophocles Mavroeidis This is my paper

Pith reviewed 2026-05-10 18:15 UTC · model grok-4.3

classification 💰 econ.EM math.STstat.TH
keywords SVARnonlinear identificationendogenous nonlinearityorthogonal transformationpiecewise affinesmooth transitionPhillips curveregime switching
0
0 comments X

The pith

SVARs with endogenous nonlinearity on the left-hand side identify parameters and shocks nonparametrically up to orthogonal transformation, under weak conditions, exactly as linear SVARs do.

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

The paper establishes that when endogenous variables enter nonlinearly in a structural vector autoregression, the model parameters and the structural shocks remain identified up to an orthogonal transformation, provided only mild regularity conditions hold on the functions and the data process. This equivalence means that the familiar identification restrictions used in linear SVARs carry over unchanged, without any increase in the number required for exact identification. Researchers can therefore model features such as asymmetric impact responses, endogenous regime switches, or occasionally binding constraints while retaining the same tools for recovering causal effects. The result is illustrated by specializing to piecewise affine and smooth-transition SVARs and by an empirical application to a nonlinear Phillips curve that tests for state dependence in inflation dynamics.

Core claim

Under weak regularity conditions the parameters and structural shocks of an endogenously nonlinear SVAR are nonparametrically identified up to an orthogonal transformation, exactly as in the linear case. Consequently most existing linear identification schemes apply directly, and the number of restrictions needed for exact identification is unchanged. The argument covers piecewise affine SVARs for endogenous regime switching and their smooth-transition counterparts, and it is used to test for nonlinearity in the Phillips curve in a manner robust to the choice of identifying assumptions.

What carries the argument

Nonparametric identification up to orthogonal transformation, which shows that endogenous nonlinearity does not add new identification requirements beyond those already needed in linear SVARs.

If this is right

  • Standard linear SVAR identification schemes, such as sign restrictions or external instruments, apply directly to the nonlinear setting without modification.
  • The same number of restrictions is needed to achieve exact identification as in the linear case.
  • Piecewise affine and smooth-transition SVARs inherit the identification result and can be used for regime-switching analysis.
  • Tests for the presence of nonlinearity, such as in the Phillips curve, can be conducted while remaining robust to the specific identifying assumptions chosen.

Where Pith is reading between the lines

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

  • The result suggests that many existing nonlinear time-series specifications in macroeconomics may be identified with little additional effort once the regularity conditions are verified.
  • Applied researchers could now estimate policy-relevant nonlinear models, such as those with occasionally binding constraints, using the same software routines developed for linear SVARs.
  • The finding raises the possibility that other forms of nonlinearity, such as those appearing only on the right-hand side, might be handled by similar arguments if the left-hand side remains linear.

Load-bearing premise

Mild regularity conditions on the nonlinear functions and the underlying data-generating process suffice for the identification argument.

What would settle it

A data-generating process satisfying the model but violating the regularity conditions in which the structural shocks cannot be recovered up to orthogonal transformation from the reduced form.

Figures

Figures reproduced from arXiv: 2604.07718 by James A. Duffy, Sophocles Mavroeidis.

Figure 3.1
Figure 3.1. Figure 3.1: Smooth transitions and invertibility 4 Application: a nonlinear Phillips curve? 4.1 Formulation as an endogenous regime-switching SVAR The inflation surge that followed the COVID-19 pandemic reignited academic interest in the possibility of nonlinearity in the transmission of supply shocks to inflation. However, views on the relevance of nonlinearity are divided. On the one hand, Ball et al. (2022) and B… view at source ↗
Figure 4.1
Figure 4.1. Figure 4.1: Nonlinear Phillips curve and state-dependent multipliers [PITH_FULL_IMAGE:figures/full_fig_p023_4_1.png] view at source ↗
read the original abstract

We study identification in structural vector autoregressions (SVARs) in which the endogenous variables enter nonlinearly on the left-hand side of the model, a feature we term endogenous nonlinearity, to distinguish it from the more familiar case in which nonlinearity arises only through exogenous or predetermined variables. This class of models accommodates asymmetric impact multipliers, endogenous regime switching, and occasionally binding constraints. We show that, under weak regularity conditions, the model parameters and structural shocks are (nonparametrically) identified up to an orthogonal transformation, exactly as in a linear SVAR. Our results have the powerful implication that most existing identification schemes for linear SVARs extend directly to our nonlinear setting, with the number of restrictions required to achieve exact identification remaining unchanged. We specialise our results to piecewise affine SVARs, which provide a convenient framework for the modelling of endogenous regime switching, and their smooth transition counterparts. We illustrate our methodology with an application to the nonlinear Phillips curve, providing a test for the presence of nonlinearity that is robust to the choice of identifying assumptions, and finding significant evidence for state-dependent inflation dynamics.

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 paper claims that in SVARs featuring endogenous nonlinearity (nonlinear functions of the endogenous variables on the LHS), the structural parameters and shocks are nonparametrically identified up to an orthogonal transformation under weak regularity conditions, exactly as in linear SVARs. This implies that standard identification schemes extend directly without requiring additional restrictions. The authors specialize the result to piecewise-affine SVARs (for endogenous regime switching) and smooth-transition variants, then apply the framework to a nonlinear Phillips curve to test for state-dependent inflation dynamics.

Significance. If the identification result holds, it is significant for empirical macroeconomics. Nonlinear SVARs are used to capture asymmetries, regime switches, and occasionally binding constraints, yet identification has often been viewed as more demanding than in the linear case. Demonstrating that the problem reduces to the familiar orthogonal-transformation ambiguity lowers the barrier to these models and permits direct reuse of sign restrictions, external instruments, and other schemes. The piecewise-affine specialization and the robust nonlinearity test in the Phillips-curve application add immediate practical value.

minor comments (3)
  1. [Abstract and §3] The abstract states that 'the number of restrictions required to achieve exact identification remaining unchanged,' but the main text would benefit from an explicit statement (perhaps in §3) of how many restrictions are needed in the nonlinear case relative to the linear benchmark.
  2. [§5] In the application (§5), the claim that the nonlinearity test is 'robust to the choice of identifying assumptions' is important; a short table or set of figures comparing the test statistic across two or three distinct identifying schemes would strengthen the presentation.
  3. [§2–3] The regularity conditions enabling the nonparametric argument (invertibility, continuity, or monotonicity of the nonlinear mapping) are described as 'weak,' but a concise enumerated list or reference to a standard assumption set in §2 or §3 would help readers verify applicability to their own models.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their positive assessment of the manuscript and for recommending minor revision. The referee's summary accurately captures our central contribution: that endogenously nonlinear SVARs are nonparametrically identified up to orthogonal transformation under weak regularity conditions, so that standard linear identification schemes carry over directly without additional restrictions.

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper advances a nonparametric identification result for SVARs with endogenous nonlinearity on the LHS, showing that model parameters and shocks are identified up to orthogonal transformation under weak regularity conditions on the nonlinear functions and DGP. This extends the standard linear SVAR result directly via the reduced-form distribution pinning down the structural mapping, without any reduction to fitted parameters, self-definitional loops, or load-bearing self-citations. The specialization to piecewise-affine and smooth-transition cases follows as corollaries from the general theorem, and the Phillips curve application is illustrative rather than foundational. The derivation chain is self-contained against external benchmarks and does not invoke uniqueness theorems or ansatzes from the authors' prior work.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The result rests on standard SVAR orthogonality of shocks plus new regularity conditions for the nonlinear mapping; no free parameters are introduced in the identification claim itself.

axioms (1)
  • domain assumption Weak regularity conditions on the nonlinear functions and the data-generating process
    Invoked to guarantee nonparametric identification up to orthogonal transformation.

pith-pipeline@v0.9.0 · 5496 in / 1208 out tokens · 36911 ms · 2026-05-10T18:15:37.642035+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

45 extracted references · 45 canonical work pages

  1. [1]

    Arias, J. E., J. F. Rubio-Ramirez, and D. F. Waggoner (2018): Inference based on structural vector autoregressions identified with sign and zero restrictions: theory and applications, Econometrica, 86, 685--720

    work page 2018

  2. [2]

    Aruoba, S. B., M. Mlikota, F. Schorfheide, and S. Villalvazo (2022): SVARs with occasionally-binding constraints, Journal of Econometrics, 231, 477--499

    work page 2022

  3. [3]

    Auerbach, A. J. and Y. Gorodnichenko (2012): Measuring the output responses to fiscal policy, American Economic Journal: Economic Policy, 4, 1--27

    work page 2012

  4. [4]

    Leigh, and P

    Ball, L., D. Leigh, and P. Mishra (2022): Understanding US inflation during the COVID-19 era, Brookings Papers on Economic Activity, 2022, 1--80

    work page 2022

  5. [5]

    (2010): Building a composite help-wanted index, Economics Letters, 109, 175--178

    Barnichon, R. (2010): Building a composite help-wanted index, Economics Letters, 109, 175--178

    work page 2010

  6. [6]

    Hou, and F

    Beaudry, P., C. Hou, and F. Portier (2025): On the fragility of the nonlinear Phillips curve view of recent inflation, National Bureau of Economic Research, Working Paper 33522

    work page 2025

  7. [7]

    Benigno, P. and G. B. Eggertsson (2023): It's baaack: The surge in inflation in the 2020s and the return of the non-linear Phillips curve, National Bureau of Economic Research, Working Paper 31197

    work page 2023

  8. [8]

    Berry, S. T. and P. A. Haile (2018): Identification of nonparametric simultaneous equations models with a residual index structure, Econometrica, 86, 289--315

    work page 2018

  9. [9]

    Bingham, N. H. (2001): Random walk and fluctuation theory, Handbook of Statistics, 19, 171--213

    work page 2001

  10. [10]

    Bruns, M. and M. Piffer (2024): Tractable Bayesian estimation of smooth transition vector autoregressive models, Econometrics Journal, 27, 343--361

    work page 2024

  11. [11]

    Castelnuovo, V

    Caggiano, G., E. Castelnuovo, V. Colombo, and G. Nodari (2015): Estimating fiscal multipliers: News from a non-linear world, Economic Journal, 125, 746--776

    work page 2015

  12. [12]

    Carriero, A., T. E. Clark, M. Marcellino, and E. Mertens (2025): Forecasting with shadow rate VARs, Quantitative Economics, 16, 795--822

    work page 2025

  13. [13]

    Chan, K. S. , ed. (2009): Exploration of a Nonlinear World: an appreciation of Howell Tong's contributions to statistics , World Scientific

    work page 2009

  14. [14]

    Galichon, M

    Chernozhukov, V., A. Galichon, M. Henry, and B. Pass (2021): Identification of hedonic equilibrium and nonseparable simultaneous equations, Journal of Political Economy, 129, 842--870

    work page 2021

  15. [15]

    (1985): Nonlinear Functional Analysis, Springer

    Deimling, K. (1985): Nonlinear Functional Analysis, Springer

    work page 1985

  16. [16]

    Duffy, J. A. and S. Mavroeidis (2024): Common trends and long-run identification in nonlinear structural VARs , arXiv:2404.05349

    work page arXiv 2024

  17. [17]

    Duffy, J. A., S. Mavroeidis, and S. Wycherley (2023): Stationarity with Occasionally Binding Constraints, arXiv:2307.06190

    work page arXiv 2023

  18. [18]

    Evans, L. C. and R. F. Gariepy (2015): Measure Theory and Fine Properties of Functions, CRC Press, revised ed

    work page 2015

  19. [19]

    (1998): The robustness of identified VAR conclusions about money, Carnegie-Rochester Conference Series on Public Policy, 49, 207--244

    Faust, J. (1998): The robustness of identified VAR conclusions about money, Carnegie-Rochester Conference Series on Public Policy, 49, 207--244

    work page 1998

  20. [20]

    Friesecke, G., R. D. James, and S. M \"u ller (2002): A theorem on geometric rigidity and the derivation of nonlinear plate theory from three-dimensional elasticity, Communications on Pure and Applied Mathematics, 55, 1461--1506

    work page 2002

  21. [21]

    Gao, J. and P. C. B. Phillips (2013): Semiparametric estimation in triangular system equations with nonstationarity, Journal of Econometrics, 176, 59--79

    work page 2013

  22. [22]

    Gouri \'e roux, C., J. J. Laffont, and A. Monfort (1980): Coherency conditions in simultaneous linear equation models with endogenous switching regimes, Econometrica, 48, 675--695

    work page 1980

  23. [23]

    Monfort, and J.-P

    Gouri \'e roux, C., A. Monfort, and J.-P. Renne (2020): Identification and estimation in non-fundamental structural VARMA models, Review of Economic Studies, 87, 1915--1953

    work page 2020

  24. [24]

    Hamilton, J. D. (1994): Time Series Analysis, Princeton University Press

    work page 1994

  25. [25]

    Horn, R. A. and C. R. Johnson (2013): Matrix Analysis, C.U.P., 2nd ed

    work page 2013

  26. [26]

    Hubrich, K. and T. Ter \"a svirta (2013): Thresholds and smooth transitions in vector autoregressive models, in VAR Models in Macroeconomics -- New Developments and Applications: essays in honor of Christopher A. Sims

    work page 2013

  27. [27]

    Ikeda, D., S. Li, S. Mavroeidis, and F. Zanetti (2024): Testing the effectiveness of unconventional monetary policy in Japan and the United States, American Economic Journal: Macroeconomics, 16, 250--286

    work page 2024

  28. [28]

    (1961): Rotation and strain, Communications on Pure and Applied Mathematics, 14, 391--413

    John, F. (1961): Rotation and strain, Communications on Pure and Applied Mathematics, 14, 391--413

    work page 1961

  29. [29]

    Kilian, L. and H. L \"u tkepohl (2017): Structural Vector Autoregressive Analysis, C.U.P

    work page 2017

  30. [30]

    L \"u tkepohl, and K

    Lanne, M., H. L \"u tkepohl, and K. Maciejowska (2010): Structural vector autoregressions with Markov switching, Journal of Economic Dynamics and Control, 34, 121--131

    work page 2010

  31. [31]

    Meitz, and P

    Lanne, M., M. Meitz, and P. Saikkonen (2017): Identification and estimation of non-Gaussian structural vector autoregressions, Journal of Econometrics, 196, 288--304

    work page 2017

  32. [32]

    (2007): New Introduction to Multiple Time Series Analysis, Springer, 2nd ed

    L \"u tkepohl, H. (2007): New Introduction to Multiple Time Series Analysis, Springer, 2nd ed

    work page 2007

  33. [33]

    Matzkin, R. L. (2008): Identification in nonparametric simultaneous equations models, Econometrica, 76, 945--978

    work page 2008

  34. [34]

    --- -.1pt --- -.1pt --- (2015): Estimation of nonparametric models with simultaneity, Econometrica, 83, 1--66

    work page 2015

  35. [35]

    (2021): Identification at the zero lower bound, Econometrica, 89, 2855--2885

    Mavroeidis, S. (2021): Identification at the zero lower bound, Econometrica, 89, 2855--2885

    work page 2021

  36. [36]

    Phillips, A. W. (1958): The relation between unemployment and the rate of change of money wage rates in the United Kingdom, 1861-1957, Economica, 25, 283--299

    work page 1958

  37. [37]

    Rubio-Ramirez, J. F., D. F. Waggoner, and T. Zha (2005): Markov-switching structural vector autoregressions: theory and application, Working Paper 2005-27

    work page 2005

  38. [38]

    --- -.1pt --- -.1pt --- (2010): Structural vector autoregressions: theory of identification and algorithms for inference, Review of Economic Studies, 77, 665--696

    work page 2010

  39. [39]

    (2012): Introduction to Piecewise Differentiable Equations, Springer

    Scholtes, S. (2012): Introduction to Piecewise Differentiable Equations, Springer

    work page 2012

  40. [40]

    Sims, C. A. (1980): Macroeconomics and reality, Econometrica, 48, 1--48

    work page 1980

  41. [41]

    Sims, C. A. and T. Zha (2006): Were there regime switches in US monetary policy? American Economic Review, 96, 54--81

    work page 2006

  42. [42]

    Stock, J. H. and M. W. Watson (2018): Identification and estimation of dynamic causal effects in macroeconomics using external instruments, Economic Journal, 128, 917--948

    work page 2018

  43. [43]

    Tj stheim, and C

    Ter \"a svirta, T., D. Tj stheim, and C. W. J. Granger (2010): Modelling Nonlinear Economic Time Series, O.U.P

    work page 2010

  44. [44]

    (1998): The robustness of identified VAR conclusions about money: a comment, Carnegie-Rochester Conference Series on Public Policy, 49, 245--263

    Uhlig, H. (1998): The robustness of identified VAR conclusions about money: a comment, Carnegie-Rochester Conference Series on Public Policy, 49, 245--263

    work page 1998

  45. [45]

    --- -.1pt --- -.1pt --- (2005): What are the effects of monetary policy on output? Results from an agnostic identification procedure, Journal of Monetary Economics, 52, 381--419

    work page 2005