pith. machine review for the scientific record. sign in

arxiv: 2604.18778 · v2 · submitted 2026-04-20 · 💰 econ.EM

Recognition: unknown

Clustered Local Projections for Time-Varying Models

Ana Maria Herrera , Elena Pesavento , Alessia Scudiero
Authors on Pith no claims yet

Pith reviewed 2026-05-10 02:58 UTC · model grok-4.3

classification 💰 econ.EM
keywords clustered local projectionstime-varying parametersimpulse response functionsk-means clusteringmonetary policy uncertaintyTreasury yieldsGMM estimation
0
0 comments X

The pith

Clustered local projections recover the conditional average response to shocks when driving variables are exogenous and a weighted average of marginal effects when endogenous.

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

The paper proposes clustered local projections to estimate impulse responses in time-varying models where parameters shift according to a low-dimensional set of observables. By first grouping observations via k-means clustering and then estimating responses within clusters, the method delivers the conditional average response under exogeneity and a weighted average of conditional marginal effects under endogeneity. Monte Carlo evidence shows the approximation works in controlled settings, and the approach is applied to trace how macroeconomic versus monetary policy uncertainty shapes the path from a contractionary monetary shock to US Treasury yields. If the clustering step faithfully sorts the data, researchers gain a practical way to handle parameter instability without specifying the full functional form of time variation.

Core claim

In a class of time-varying models, the clustered LP procedure recovers the conditional average response function when the driving variables are exogenous and a weighted average of the conditional marginal effects when they are endogenous. The procedure classifies observations by k-means on the low-dimensional observable matrix, estimates cluster-specific impulse responses by GMM, and compares the resulting functions across clusters.

What carries the argument

k-means clustering on the low-dimensional matrix of observables that governs parameter variation, followed by GMM estimation of impulse responses inside each resulting group.

If this is right

  • When the driving variables are exogenous, the clustered LP estimates equal the conditional average response.
  • When the driving variables are endogenous, the clustered LP estimates equal a weighted average of the conditional marginal effects.
  • The method isolates distinct channels through which macroeconomic uncertainty and monetary policy uncertainty affect the term premium and expectations revisions after a monetary shock.
  • Iterative application of clustering, GMM estimation, and cross-cluster comparison yields usable impulse responses even when full time-varying parameter models are difficult to estimate.

Where Pith is reading between the lines

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

  • The same clustering step could be applied to other endogenous regressors whose effects are suspected to vary with observable states, such as fiscal multipliers conditional on debt levels.
  • If the number of clusters is chosen by cross-validation rather than fixed in advance, the procedure might reduce sensitivity to the arbitrary choice of k.
  • Extending the framework to allow cluster membership to depend on lagged outcomes would address cases where the grouping variable itself responds to the shock.

Load-bearing premise

The parameter variation in the time-varying model is linked to a low-dimensional matrix of observables that can be accurately captured by k-means clustering into discrete groups.

What would settle it

A Monte Carlo experiment in which the true data-generating process has parameter variation driven by a high-dimensional or smoothly varying process outside the span of k-means clusters, with the clustered LP estimates then failing to track the known conditional average responses.

Figures

Figures reproduced from arXiv: 2604.18778 by Alessia Scudiero, Ana Maria Herrera, Elena Pesavento.

Figure 1
Figure 1. Figure 1: Univariate Threshold Model -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 0 5 10 15 Horizon -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 0 5 10 15 Horizon -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 0 5 10 15 Horizon -4 -2 0 2 4 Notes: This figure illustrates the true impulse response functions for each Zt−1 in gray, the clustered LP estimate in (purple), and the CARh (δ = 1… view at source ↗
Figure 2
Figure 2. Figure 2: Bivariate Threshold Model -4 -2 0 2 4 -4 -2 0 2 4 0 5 10 15 Horizon -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 0 5 10 15 Horizon -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 -4 -2 0 2 4 0 5 10 15 Horizon -4 -2 0 2 4 Notes: This figure illustrates the true impulse response functions for each Zt−1 in gray, the clustered LP estimate in (purple), and the CARh (δ = 1, k) (in red) for each partition (K =… view at source ↗
Figure 3
Figure 3. Figure 3: Absolute Value Model -0.5 0 0.5 1 0 5 10 15 Horizon -0.5 0 0.5 1 -0.5 0 0.5 1 -0.5 0 0.5 1 0 5 10 15 Horizon -0.5 0 0.5 1 -0.5 0 0.5 1 -0.5 0 0.5 1 -0.5 0 0.5 1 0 5 10 15 Horizon -0.5 0 0.5 1 Notes: This figure illustrates the true impulse response functions for each Zt−1 in gray, the clustered LP estimate in (purple), and the CARh (δ = 1, k) (in red) for each partition (K = 2 in the left panel, K = 3 in t… view at source ↗
Figure 7
Figure 7. Figure 7: Classification of Macroeconomic and Monetary Policy Uncertainty Indices [PITH_FULL_IMAGE:figures/full_fig_p027_7.png] view at source ↗
read the original abstract

We propose a clustered local projection (clustered LP) method to estimate impulse response functions in a class of time-varying models where parameter variation is linked to a low-dimensional matrix of observables. We show that the clustered LP recovers the conditional average response when the driving variables are exogenous and a weighted average of the conditional marginal effects when they are endogenous. We propose an iterative estimation method that first classifies the data using k-means, estimates impulse response functions via GMM, and evaluates differences across clustered LP estimates. Our Monte Carlo simulations illustrate the ability of clustered LP to approximate the conditional average response function. We employ our technique to examine how uncertainty influences the transmission of a contractionary monetary policy shock to the 5- and 10-year U.S. nominal Treasury yields. Our estimation results suggest macroeconomic and monetary policy uncertainty operate through complementary but distinct channels: the former primarily amplifies the risk compensation embedded in the term premium, while the latter governs the speed and persistence with which markets revise their expectations about the future rate path following a monetary policy shock.

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

2 major / 1 minor

Summary. The paper proposes a clustered local projection (clustered LP) estimator for impulse response functions in time-varying parameter models where parameter variation is linked to a low-dimensional matrix of observables. It claims that k-means clustering followed by GMM estimation of local projections within clusters recovers the conditional average response when driving variables are exogenous and a weighted average of conditional marginal effects when endogenous. Monte Carlo simulations are presented to illustrate approximation of the conditional response, and the method is applied to study how macroeconomic and monetary policy uncertainty affect the transmission of contractionary monetary policy shocks to 5- and 10-year U.S. Treasury yields.

Significance. If the clustering step reliably isolates regimes of approximately constant parameters, the method offers a computationally tractable way to estimate state-dependent impulse responses without fully nonparametric or regime-switching specifications. The Monte Carlo evidence provides illustrative support for the approximation property, and the empirical application yields interpretable distinctions between uncertainty channels. This could be a useful addition to the toolkit for macroeconometric analysis of nonlinear policy effects.

major comments (2)
  1. [Theoretical results] Theoretical results section: The recovery property is derived under the maintained assumption that k-means clustering on the observable matrix produces groups in which the time-varying parameters are approximately constant. No conditions on cluster separability, bounded misclassification probability, or properties of the observables-to-parameter mapping are provided. When this mapping is continuous or noisy, k-means can mix regimes, so the GMM estimates converge to a mixture rather than the claimed conditional quantity; this assumption is load-bearing for the central claim.
  2. [Monte Carlo simulations] Monte Carlo section: The simulations are reported to support approximation of the conditional average response, but no diagnostics on cluster recovery (e.g., misclassification rates, adjusted Rand index, or purity metrics) or sensitivity to the free parameter (number of clusters) and k-means initialization are included. Without these, finite-sample performance of the recovery cannot be assessed, undermining verification of the method's reliability.
minor comments (1)
  1. [Abstract] The abstract states an 'iterative estimation method' involving classification, GMM, and evaluation of differences, but the precise iteration (e.g., whether clustering is updated after initial GMM estimates) is not detailed; a brief clarification would improve reproducibility.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and insightful comments on our paper. We address the major comments point by point below and outline the revisions we plan to make to strengthen the manuscript.

read point-by-point responses

  1. Referee: [Theoretical results] Theoretical results section: The recovery property is derived under the maintained assumption that k-means clustering on the observable matrix produces groups in which the time-varying parameters are approximately constant. No conditions on cluster separability, bounded misclassification probability, or properties of the observables-to-parameter mapping are provided. When this mapping is continuous or noisy, k-means can mix regimes, so the GMM estimates converge to a mixture rather than the claimed conditional quantity; this assumption is load-bearing for the central claim.

    Authors: We agree that the central recovery result relies on the clustering procedure producing groups with approximately constant parameters. The manuscript explicitly maintains this assumption, as stated in the theoretical section, where we derive that under this condition the clustered LP recovers the conditional average response for exogenous drivers. We do not provide primitive conditions on the observables-to-parameter mapping or k-means performance because the focus is on the estimation procedure conditional on successful clustering. However, we acknowledge the referee's point and will revise the paper to include a more detailed discussion of when this assumption is likely to hold, such as when the low-dimensional observables exhibit regime-like behavior, and note the potential for mixing in continuous cases. This will clarify the scope of the method without altering the main claims. revision: partial

  2. Referee: [Monte Carlo simulations] Monte Carlo section: The simulations are reported to support approximation of the conditional average response, but no diagnostics on cluster recovery (e.g., misclassification rates, adjusted Rand index, or purity metrics) or sensitivity to the free parameter (number of clusters) and k-means initialization are included. Without these, finite-sample performance of the recovery cannot be assessed, undermining verification of the method's reliability.

    Authors: We appreciate this suggestion for improving the Monte Carlo analysis. In the revised manuscript, we will augment the simulation section with diagnostics on cluster recovery, including misclassification rates and adjusted Rand indices across different data generating processes. We will also report results for varying numbers of clusters and multiple k-means initializations to demonstrate robustness. These additions will provide a more complete assessment of the method's finite-sample properties and support the approximation claims. revision: yes

Circularity Check

0 steps flagged

No circularity: recovery claims are derived under explicit clustering assumptions without tautological reduction

full rationale

The paper's central result states that clustered LP recovers the conditional average response (exogenous case) or weighted marginal effects (endogenous case) when parameters are linked to observables accurately captured by k-means. This is presented as a derived property of the GMM estimator applied to the resulting clusters, not as a fit to the target quantity itself. Monte Carlo results are described as illustrating approximation rather than exact recovery by construction. No self-citation chains, fitted parameters renamed as predictions, or ansatzes smuggled via prior work appear in the provided text. The derivation chain remains self-contained against the stated low-dimensional observable assumption and does not reduce to its inputs.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The approach rests on standard econometric assumptions for local projections and GMM consistency, plus the domain assumption that parameter heterogeneity is captured by low-dimensional observables amenable to k-means clustering. No new entities are postulated.

free parameters (1)
  • number of clusters
    Chosen for k-means classification; value not specified in abstract but affects grouping of time periods.
axioms (2)
  • domain assumption Driving variables are either exogenous or the weighting in the average marginal effect is well-defined under endogeneity.
    Invoked to state the recovery properties of clustered LP.
  • domain assumption k-means clustering accurately partitions the data according to the low-dimensional observable matrix.
    Central to the iterative estimation procedure described.

pith-pipeline@v0.9.0 · 5477 in / 1460 out tokens · 33262 ms · 2026-05-10T02:58:03.898992+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

99 extracted references · 13 canonical work pages

  1. [1]

    Sims , title =

    C. Sims , title =. Federal Reserve Bank of Minneapolis, Quarterly Review (Winter) , year =

  2. [2]

    Auerbach and Y

    A. Auerbach and Y. Gorodnichenko , title =. 2012 , journal =

    2012

  3. [3]

    Auerbach and Y

    A. Auerbach and Y. Gorodnichenko. , title =. Alesina, A., Giavazzi, F. (eds.): Fiscal Policy after the Financial Crisis. University of Chicago Press , year =

  4. [4]

    Auerbach and Y.Gorodnichenko , title =

    A. Auerbach and Y.Gorodnichenko , title =. American Economic Review Papers and Proceedings 103 , pages =

  5. [5]

    and Sims, E.R , title =

    Bachmann, R. and Sims, E.R , title =. Journal of Monetary Economics 59 , pages =

  6. [6]

    and Ramey, V.A

    Owyang, M.T. and Ramey, V.A. and Zubairy, S. , title =. American Economic Review: Papers and Proceedings 103 , year =

  7. [7]

    Ramey, V. A. and Zubairy, S. , title =. Journal of Political Economy 126 , year =

  8. [8]

    and Zanetti, F

    Ghassibe, M. and Zanetti, F. , title =. Manuscript, University of Oxford. , year =

  9. [9]

    , title =

    Alloza, M. , title =. Manuscript, Bank of Spain , year =

  10. [10]

    and Lieb, L

    Candelon, B. and Lieb, L. , title =. Journal of Economic Dynamics and Control 37 , year =

  11. [11]

    and Azzalini, G

    Alesina, A. and Azzalini, G. and Favero, C. and Giavazzi, F. and Miano, A. , title =. IMF Economic Review 66 , year =

  12. [12]

    , title =

    Eskandari, R. , title =. Manuscript, University of York , year =

  13. [13]

    , title =

    Demirel, U.D. , title =. Journal of Monetary Economics 117 , year =

  14. [14]

    2024a , doi =

    State-dependent local projections , journal =. 2024a , doi =

  15. [15]

    2024b , author =

    Nonparametric Local Projections\ , journal =. 2024b , author =

  16. [16]

    and Jordà, Ò

    Cloyne, J. and Jordà, Ò. and Taylor, A. M. , title =. Working Paper 30971 http://www.nber.org/papers/w30971 , year =

  17. [17]

    , title =

    Potter, S. , title =. Journal of Economic Dynamics and Control, vol. 24 issue 10 , year =

  18. [18]

    and Matthes, C

    Lubik, T.A. and Matthes, C. , title =. Economic Quarterly Volume 101, Number 4, Fourth Quarter , year =

  19. [19]

    and Rossi, B

    Inoue, A. and Rossi, B. and Wang, Y. , title =. CEPR Discussion Paper No. DP17134 , year =

  20. [20]

    , title =

    Primiceri, G. , title =. Review of Economic Studies, 72 , year =

  21. [21]

    and Petrella, I

    Santoro, E. and Petrella, I. and Pfajfar, D. and Gaffeo, E. , title =. Journal of Monetary Economics 68 , year =

  22. [22]

    and Thwaites, G

    Tenreyro, S. and Thwaites, G. , title =. American Economic Journal: Macroeconomics 8 , year =

  23. [23]

    and Jordà, O

    Angrist, J.D. and Jordà, O. and Kuersteiner, G.M. , title =. Journal of Business and Economic Statistics 36 , year =

  24. [24]

    and Clark, T.E

    Carriero, A. and Clark, T.E. and Marcellino, M. , title =. The Review of Economics and Statistics 2018; 100 (5) , year =

    2018

  25. [25]

    and Bai, R

    Lin, B. and Bai, R. , title =. Research in International Business and Finance, Volume 56 , year =

  26. [26]

    and Ellison, M

    Ehrmann, M. and Ellison, M. and Valla, N. , title =. Economics Letters, Volume 56 , year =

  27. [27]

    2025 , note =

    Maung, Kenwin and Yoshida, Nathaniel , title =. 2025 , note =

    2025

  28. [28]

    , title =

    Pascal, P. , title =. The Review of Economics and Statistics , year =

  29. [29]

    and Mumtaz, H

    Kapetanios, G. and Mumtaz, H. and Stevens, I. and Theodoridis, K. , title =. The Economic Journal , year =

  30. [30]

    and Gambetti, L

    Galí, J. and Gambetti, L. , title =. American Economic Journal: Macroeconomics , year =

  31. [31]

    and Min, F

    Wen, F. and Min, F. and Zhang, Y-J. and Yang, C. , title =. International Journal of Finance and Economics , year =

  32. [32]

    and Petalas, P-E

    Müller, U.K. and Petalas, P-E. , title =. The Review of Economic Studies , year =

  33. [33]

    and Shephard, N

    Rambachan, A. and Shephard, N. , title =. Working Paper , year =

  34. [34]

    and Rossi, P.E

    Gallant, A.R. and Rossi, P.E. and Tauchen, G. , title =. Econometrica , year =

  35. [35]

    and Pesaran, H

    Koop, G. and Pesaran, H. and Potter, S. , title =. Journal of Econometrics , year =

  36. [36]

    and Uzeda, L

    Leiva-León, D. and Uzeda, L. , title =. Working Paper, Bank of Canada , year =

  37. [37]

    and Uzeda, L

    Leiva-León, D. and Uzeda, L. , title = ". The Review of Economics and Statistics , volume =. 2023 , month =

    2023

  38. [38]

    and Méa, A

    Séraphin, P. and Méa, A. , year =. Bank Credit Channel of Monetary Policy in the UEMOA Zone: A TVP-VAR Approach , volume =. International Journal of Economics and Finance , doi =

  39. [39]

    , Title =

    Jordà, Ò. , Title =. American Economic Review , Volume =. 2005 , Month =

    2005

  40. [40]

    , journal =

    Jordà, Ò. , journal =. Simoultaneous confidence regions for impulse responses , volume =

  41. [41]

    and Peersman, G

    Baumeister, C. and Peersman, G. , Title =. American Economic Journal: Macroeconomics , Volume =. 2013 , Month =

    2013

  42. [42]

    and Gambetti, L

    Canova, F. and Gambetti, L. , title =. Journal of Economic Dynamics and Control , volume =. 2009 , issn =

    2009

  43. [43]

    2011 , journal =

    Time-Varying Parameter VAR Model with Stochastic Volatility: An Overview of Methodology and Empirical Applications , author =. 2011 , journal =

    2011

  44. [44]

    2021 , note =

    Impulse response analysis for structural dynamic models with nonlinear regressors , journal =. 2021 , note =. doi:https://doi.org/10.1016/j.jeconom.2021.06.009 , author =

    work page doi:10.1016/j.jeconom.2021.06.009 2021

  45. [45]

    Gourieroux and J

    C. Gourieroux and J. Jasiak , journal =. Nonlinear Innovations and Impulse Responses with Application to VaR Sensitivity , year =

  46. [47]

    De Santis, R.A. and P. Surico and M.B. Devereux and D. Giannone , journal =. Bank lending and monetary transmission in the euro area , volume =

  47. [48]

    2005 , journal =

    Inside the bank lending channel , author =. 2005 , journal =

    2005

  48. [49]

    2011 , journal =

    The bank lending channel: lessons from the crisis , author =. 2011 , journal =

    2011

  49. [50]

    2022 , journal =

    Money Markets and Bank Lending: Evidence from the Adoption of Tiering , author =. 2022 , journal =

    2022

  50. [51]

    and Boucinha, M

    Altavilla, C. and Boucinha, M. and Holton, S. and Ongena, S. , title =. Journal of Money, Credit and Banking , volume =. doi:https://doi.org/10.1111/jmcb.12792 , year =

    work page doi:10.1111/jmcb.12792

  51. [52]

    Economics and Business Letters , author=

    Transmission mechanisms in Japan’s quantitative easing policy (2001–2006) , volume=. Economics and Business Letters , author=. 2017 , pages=. doi:10.17811/ebl.6.2.2017.35-41 , number=

    work page doi:10.17811/ebl.6.2.2017.35-41 2001

  52. [53]

    Journal of Money, Credit and Banking , volume =

    Jo, Soojin , title =. Journal of Money, Credit and Banking , volume =

  53. [54]

    Price Setting and Volatility: Evidence from Oil Price Volatility Shocks , month =

    Klepacz, Matthew , journal =. Price Setting and Volatility: Evidence from Oil Price Volatility Shocks , month =

  54. [55]

    Journal of Monetary Economics , volume =

    Disagreement about inflation expectations and monetary policy transmission , author =. Journal of Monetary Economics , volume =. 2021 , issn =

    2021

  55. [56]

    Monetary Policy , author =

    The Global Transmission of U.S. Monetary Policy , author =. Working paper , year =

  56. [57]

    NBER Macroecon

    Does monetary policy matter? A new test in the spirit of Friedman and Schwartz , author =. NBER Macroecon. Annu. , year =

  57. [58]

    Carnegie-Rochester Conference Series on Public Policy , year =

    Costly capital reallocation and the effects of government spending , author =. Carnegie-Rochester Conference Series on Public Policy , year =

  58. [59]

    The Quarterly Journal of Economics , year =

    Identifying government spending shocks: It’s all in the timing , author =. The Quarterly Journal of Economics , year =

  59. [60]

    Handbook of Macroeconomics , year =

    Macroeconomic shocks and their propagation , author =. Handbook of Macroeconomics , year =

  60. [61]

    Monetary policy uncertainty and monetary policy surprises , journal =

    Michiel. Monetary policy uncertainty and monetary policy surprises , journal =. 2021 , doi =

    2021

  61. [62]

    The Economic Journal , volume =

    Bauer, Michael D and Lakdawala, Aeimit and Mueller, Philippe , title = ". The Economic Journal , volume =. 2021 , month =

    2021

  62. [64]

    and Romer, David H

    Romer, Christina D. and Romer, David H. , Title =. American Economic Review , Volume =. 2004 , Month =

    2004

  63. [65]

    and Bloom, Nick and Davis, Stephen J

    Baker, Scott R. and Bloom, Nick and Davis, Stephen J. , Year =. Economic Policy Uncertainty Index for United States - US Daily News Index , url =

  64. [66]

    2024 , issn =

    Local projections in unstable environments , journal =. 2024 , issn =. doi:https://doi.org/10.1016/j.jeconom.2024.105726 , author =

    work page doi:10.1016/j.jeconom.2024.105726 2024

  65. [67]

    2019 , author =

    Time-varying local projections , journal =. 2019 , author =

    2019

  66. [68]

    Journal of Money, Credit and Banking , volume =

    Tillman, Peter , title =. Journal of Money, Credit and Banking , volume =. doi:https://doi.org/10.1111/jmcb.12657 , year =

    work page doi:10.1111/jmcb.12657

  67. [69]

    Journal of Financial Econometrics , volume =

    Han, Heejoon and Jung, Whayoung and Lee, Ji Hyung , title = ". Journal of Financial Econometrics , volume =. 2022 , month =

    2022

  68. [70]

    and Ng, Serena , Title =

    Jurado, Kyle and Ludvigson, Sydney C. and Ng, Serena , Title =. American Economic Review , Volume =. 2015 , Month =. doi:10.1257/aer.20131193 , URL =

    work page doi:10.1257/aer.20131193 2015

  69. [71]

    and Swanson, Eric T

    Bauer, Michael D. and Swanson, Eric T. , title =. NBER Macroeconomics Annual , volume =. 2023 , doi =

    2023

  70. [72]

    and Huber,F

    Hauzenberger, N. and Huber,F. and Koop, G. and Mitchell, J. , title =. Working Paper , year =

  71. [73]

    2020 , issn =

    Reducing the state space dimension in a large TVP-VAR , journal =. 2020 , issn =. doi:https://doi.org/10.1016/j.jeconom.2019.11.006 , author =

    work page doi:10.1016/j.jeconom.2019.11.006 2020

  72. [74]

    and Hauzenberger, Niko and Huber, Florian and Pfarrhofer, Michael , title =

    Fischer, Manfred M. and Hauzenberger, Niko and Huber, Florian and Pfarrhofer, Michael , title =. Journal of Applied Econometrics , volume =. doi:https://doi.org/10.1002/jae.2936 , year =

    work page doi:10.1002/jae.2936

  73. [75]

    2003 , publisher =

    Nonlinear Time Series: Nonparametric and Parametric Methods , author =. 2003 , publisher =

    2003

  74. [76]

    Journal of the American Statistical Association , volume =

    Functional-Coefficient Autoregressive Models , author =. Journal of the American Statistical Association , volume =

  75. [77]

    Journal of the American Statistical Association , volume =

    Functional-Coefficient Regression Models for Nonlinear Time Series , author =. Journal of the American Statistical Association , volume =

  76. [78]

    Inflation disagreement weakens the power of monetary policy , author=

  77. [79]

    Journal of Business & Economic Statistics , volume =

    Michal Kolesár and Mikkel Plagborg-Møller , title =. Journal of Business & Economic Statistics , volume =. 2025 , publisher =

    2025

  78. [80]

    2025 , eprint=

    Dynamic Causal Effects in a Nonlinear World: the Good, the Bad, and the Ugly , author=. 2025 , eprint=

    2025

  79. [81]

    2024 , journal=

    Inference for Local Projections , author=. 2024 , journal=

    2024

  80. [82]

    , title =

    Kilian, Lutz and Vigfusson, Robert J. , title =. Quantitative Economics , volume =. doi:https://doi.org/10.3982/QE99 , year =

    work page doi:10.3982/qe99

Showing first 80 references.