pith. sign in

arxiv: 2606.08141 · v1 · pith:6F5H3TNRnew · submitted 2026-06-06 · 💰 econ.EM · q-fin.GN

A Structural Matrix Autoregressive Model for the Joint Dynamics of Volume, Volatility, and Returns

Andrea Bucci , Giulio Palomba , Eduardo Rossi This is my paper

Pith reviewed 2026-06-27 18:52 UTC · model grok-4.3

classification 💰 econ.EM q-fin.GN
keywords structural matrix autoregressive modeltrading volumerealized volatilityasset returnsspilloversmixture of distributions hypothesisFOMC announcementsforecast error variance decomposition
0
0 comments X

The pith

Volatility is the primary driver of trading activity in a structural model of returns, volatility and volume.

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

The paper introduces a Structural Matrix Autoregressive model that jointly analyzes asset returns, realized volatility and trading volume across many assets while remaining parsimonious. Structural restrictions drawn from the Mixture of Distributions Hypothesis and efficient market theory identify the informative component of volume. Estimation on daily Dow Jones data from 2021 to 2025 shows that volatility shocks drive trading volume, implying that informational shocks reach markets chiefly through price variability. Forecast error variance decompositions indicate internal shocks dominate short-term volume but cross-asset spillovers explain more than half the variation at longer horizons. An event study around FOMC announcements finds temporary increases in the informative volume component followed by quick reversion.

Core claim

The SMAR model, estimated on daily data for Dow Jones constituents from 2021 to 2025 and identified via restrictions consistent with the Mixture of Distributions Hypothesis and efficient market theory, shows that volatility shocks are the main source of trading-volume responses, with informational effects transmitted primarily through price variability rather than direct volume shocks.

What carries the argument

The Structural Matrix Autoregressive (SMAR) model, which imposes theory-based restrictions to separate informative volume shocks from liquidity or noise components while capturing cross-variable and cross-asset spillovers.

If this is right

  • Volatility shocks explain the largest share of trading-volume variation.
  • Cross-asset spillovers account for more than 50 percent of trading-volume variation at longer horizons.
  • Internal shocks dominate short-term volume dynamics.
  • The informative component of volume rises on FOMC announcement days and reverts rapidly afterward.

Where Pith is reading between the lines

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

  • If the identification holds, volume-based measures of information flow may be less reliable than volatility-based ones.
  • The framework could be applied to test whether the same volatility-volume ordering appears in equity markets outside the Dow Jones or in other asset classes.
  • Central banks might treat volatility spikes as early signals of subsequent trading surges when assessing market reactions to policy news.

Load-bearing premise

The theory-based restrictions correctly isolate the informative component of volume from liquidity or noise trading.

What would settle it

Finding that volume no longer responds primarily to volatility shocks after replacing the Mixture of Distributions Hypothesis restrictions with an alternative identification scheme, or observing no rise in the informative volume component on FOMC announcement days.

Figures

Figures reproduced from arXiv: 2606.08141 by Andrea Bucci, Eduardo Rossi, Giulio Palomba.

Figure 1
Figure 1. Figure 1: Average structural impulse response function Note: The light blue area represents the 90% bootstrapped confidence interval on 200 replicates. The first variable is the one shocked, while the second one is the response. 19 [PITH_FULL_IMAGE:figures/full_fig_p019_1.png] view at source ↗
Figure 3
Figure 3. Figure 3: Informative share of volume around FOMC announcements (Validated) 50% 75% 100% 125% −10 −5 0 5 10 Days relative to announcement Informative share of volume variance Note: The plot shows the average daily informative share of volume variance across 33 FOMC events. The spike at t = 0 denotes the contemporaneous impact of the macro announcement. The dotted line represents the sample mean, and the shaded area … view at source ↗
read the original abstract

This paper proposes a Structural Matrix Autoregressive (SMAR) model for the joint analysis of asset returns, realized volatility, and trading volume in a large-dimensional setting. This framework simultaneously captures dynamic spillovers across financial variables and cross-sectional dependence across assets while preserving a parsimonious parameterization relative to conventional vector autoregressive models. The model is estimated on daily data for the constituents of the Dow Jones Industrial Average over the period 2021-2025 and is structurally identified through restrictions consistent with the Mixture of Distributions Hypothesis and efficient market theory. The empirical findings indicate that volatility is the primary driver of trading activity, suggesting that informational shocks are predominantly incorporated into markets through price variability. Forecast error variance decompositions further reveal that, although internal shocks dominate short-term volume dynamics, cross-asset spillovers account for more than 50% of trading volume variation at longer horizons. Finally, an event-study analysis around FOMC announcements supports the proposed decomposition by identifying significant increases in the informative component of trading activity on announcement days followed by rapid mean reversion.

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 Structural Matrix Autoregressive (SMAR) model for the joint dynamics of asset returns, realized volatility, and trading volume in a large-dimensional panel. The model is estimated on daily data for Dow Jones Industrial Average constituents (2021-2025) and is structurally identified via restrictions drawn from the Mixture of Distributions Hypothesis and efficient-market theory. The main empirical claims are that volatility shocks are the primary driver of trading activity, that cross-asset spillovers explain more than 50% of long-horizon volume variation, and that an event study around FOMC announcements corroborates the decomposition by showing temporary increases in the informative component of volume.

Significance. If the identification is valid, the SMAR framework supplies a parsimonious way to separate informative from non-informative volume while quantifying both cross-variable and cross-asset spillovers; the resulting variance decompositions and event-study evidence would be a useful contribution to the microstructure literature on how information is incorporated into prices and trading.

major comments (2)
  1. [Abstract / Identification] Abstract and identification section: the structural restrictions are stated to be 'consistent with' MDH and efficient-market theory, yet the manuscript supplies no information on whether the restrictions are over-identifying, no over-identification tests, and no comparison with sign-restricted or heteroskedasticity-based alternatives. Because the headline claim that 'volatility is the primary driver of trading activity' is obtained only after imposing these restrictions to isolate the informative component of volume, the absence of validation is load-bearing for the central result.
  2. [Variance Decompositions] Forecast-error variance decomposition results: the claim that cross-asset spillovers account for more than 50% of trading-volume variation at longer horizons is presented without any reported lag-selection procedure, standard-error construction, or robustness checks to alternative lag lengths. This directly affects the quantitative interpretation of the spillover magnitudes.
minor comments (1)
  1. [Abstract] The abstract does not state the exact number of assets or the precise estimation sample size, which would help readers gauge the dimensionality of the panel.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address each major point below and indicate planned revisions.

read point-by-point responses

  1. Referee: [Abstract / Identification] Abstract and identification section: the structural restrictions are stated to be 'consistent with' MDH and efficient-market theory, yet the manuscript supplies no information on whether the restrictions are over-identifying, no over-identification tests, and no comparison with sign-restricted or heteroskedasticity-based alternatives. Because the headline claim that 'volatility is the primary driver of trading activity' is obtained only after imposing these restrictions to isolate the informative component of volume, the absence of validation is load-bearing for the central result.

    Authors: The restrictions are derived directly from the Mixture of Distributions Hypothesis (volume driven by information-induced volatility) and efficient-market theory (returns incorporate information immediately). These yield a just-identified system in the panel setting. We agree that explicit validation would strengthen the paper. In revision we will add a dedicated subsection on the identifying assumptions, clarify their just-identifying nature, and report robustness exercises that replace selected restrictions with sign restrictions or heteroskedasticity-based identification where feasible. Direct Sargan-type over-identification tests are not straightforward in this high-dimensional panel, but we will discuss this limitation explicitly. revision: partial

  2. Referee: [Variance Decompositions] Forecast-error variance decomposition results: the claim that cross-asset spillovers account for more than 50% of trading-volume variation at longer horizons is presented without any reported lag-selection procedure, standard-error construction, or robustness checks to alternative lag lengths. This directly affects the quantitative interpretation of the spillover magnitudes.

    Authors: The lag order was selected by panel information criteria, but these steps and associated uncertainty measures were omitted. In the revised manuscript we will document the lag-selection procedure, supply bootstrap standard errors for the FEVDs that respect the panel structure, and present robustness tables for alternative lag lengths (AIC, BIC, and fixed lags of 1–5). These additions will allow readers to assess the sensitivity of the long-horizon spillover shares. revision: yes

Circularity Check

0 steps flagged

No circularity: identification uses external MDH/EMT theory; FEVDs follow from estimated model

full rationale

The paper imposes structural restrictions drawn from the Mixture of Distributions Hypothesis and efficient market theory to identify the SMAR model. These restrictions are stated as external theoretical consistency conditions rather than being fitted to the same data or derived from the model's own outputs. The reported dominance of volatility shocks and the >50% long-horizon spillover share in volume FEVDs are direct consequences of the estimated coefficients under those restrictions; they do not reduce to the restrictions by algebraic identity or by renaming a fitted parameter. No self-citation chains, self-definitional loops, or ansatzes smuggled via prior work appear in the abstract or described methodology. The derivation chain therefore remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claims rest on two domain assumptions drawn from financial theory and on the usual maintained assumptions of a linear VAR-type model; no new entities are postulated and no parameters are described as fitted ad hoc beyond standard estimation.

axioms (2)
  • domain assumption Mixture of Distributions Hypothesis
    Invoked to justify the structural restrictions that separate informative from non-informative volume components.
  • domain assumption Efficient market theory (rapid price adjustment)
    Invoked to justify the structural restrictions used for identification.

pith-pipeline@v0.9.1-grok · 5716 in / 1598 out tokens · 18300 ms · 2026-06-27T18:52:44.626521+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

113 extracted references · 13 canonical work pages

  1. [1]

    and Pedersen, Lasse Heje , title=

    Acharya, Viral V. and Pedersen, Lasse Heje , title=. Journal of Financial Economics , year=

  2. [2]

    Journal of Financial Economics , year =

    Amihud, Yakov and Mendelson, Haim , title =. Journal of Financial Economics , year =

  3. [3]

    Journal of Financial Markets , year =

    Amihud, Yakov , title =. Journal of Financial Markets , year =

  4. [4]

    and Bollerslev, Tim and Diebold, Francis X

    Andersen, Torben G. and Bollerslev, Tim and Diebold, Francis X. and Labys, Paul , title =. The distribution of realized exchange rate volatility , year =

  5. [5]

    and Bollerslev, Tim and Diebold, Francis X

    Andersen, Torben G. and Bollerslev, Tim and Diebold, Francis X. and Ebens, Heiko , title=. Journal of Financial Economics , year=

  6. [6]

    and Bollerslev, Tim and Diebold, Francis X

    Andersen, Torben G. and Bollerslev, Tim and Diebold, Francis X. and Labys, Paul , title =. Econometrica , year =

  7. [7]

    and Bollerslev, Tim and Diebold, Francis X

    Andersen, Torben G. and Bollerslev, Tim and Diebold, Francis X. and Vega, Clara , title =. American Economic Review , year =

  8. [8]

    , title=

    Andersen, Torben G. , title=. The Journal of Finance , year=

  9. [9]

    Econometrica , year=

    Archakov, Ilya and Hansen, Peter Reinhard , title=. Econometrica , year=

  10. [10]

    SIAM Journal on Matrix Analysis and Applications , volume =

    Arsigny, Vincent and Fillard, Pierre and Pennec, Xavier and Ayache, Nicholas , title =. SIAM Journal on Matrix Analysis and Applications , volume =. 2007 , doi =

    2007

  11. [11]

    Journal of Econometrics , year=

    Asai, Manabu and McAleer, Michael , title=. Journal of Econometrics , year=

  12. [12]

    Journal of Econometrics , year=

    Asai, Manabu and Chang, Chia-Lin and McAleer, Michael , title=. Journal of Econometrics , year=

  13. [13]

    Econometrica , year=

    Bai, Jushan and Ng, Serena , title=. Econometrica , year=

  14. [14]

    and Bollerslev, Tim and Mikkelsen, Hans Ole , title=

    Baillie, Richard T. and Bollerslev, Tim and Mikkelsen, Hans Ole , title=. Journal of Econometrics , year=

  15. [15]

    Bandi, F. M. and Ren. Price and volatility co-jumps , journal=. 2016 , volume=

    2016

  16. [16]

    and Shephard, Neil , title =

    Barndorff-Nielsen, Ole E. and Shephard, Neil , title =. Journal of Applied Econometrics , year =

  17. [17]

    and Shephard, Neil , title =

    Barndorff-Nielsen, Ole E. and Shephard, Neil , title =. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , year =

  18. [18]

    and Vorkink, Keith , title =

    Bauer, Gregory H. and Vorkink, Keith , title =. Journal of Econometrics , year =

  19. [19]

    , title =

    Bessembinder, Hendrik and Seguin, Paul J. , title =. The Journal of Financial and Quantitative Analysis , year =

  20. [20]

    , title =

    Black, F. , title =. Proceedings of the 1976 Meeting of the Business and Economic Statistics Section, American Statistical Association, Washington DC , year =

    1976

  21. [21]

    Journal of Business & Economic Statistics , year =

    Bollerslev, Tim and Jubinski, Dan , title =. Journal of Business & Economic Statistics , year =

  22. [22]

    Journal of Financial Econometrics , year=

    Bollerslev, Tim and Litvinova, Julia and Tauchen, George , title=. Journal of Financial Econometrics , year=. doi:10.1093/jjfinec/nbj014 , url=

    work page doi:10.1093/jjfinec/nbj014

  23. [23]

    The Review of Economic Studies , year=

    Bollerslev, Tim and Li, Jia and Xue, Yuan , title=. The Review of Economic Studies , year=

  24. [24]

    and Quaedvlieg, Rogier , title=

    Bollerslev, Tim and Patton, Andrew J. and Quaedvlieg, Rogier , title=. Journal of Econometrics , year=

  25. [25]

    and Wu, Eliza , title=

    Brogaard, Jonathan and Nguyen, Thanh Huong and Putnins, Talis J. and Wu, Eliza , title=. The Review of Financial Studies , year=

  26. [26]

    Advances in Neural Information Processing Systems , volume=

    Brooks, Daniel and Schwander, Olivier and Barbaresco, Fr. Advances in Neural Information Processing Systems , volume=

  27. [27]

    and Pedersen, Lasse Heje , title=

    Brunnermeier, Markus K. and Pedersen, Lasse Heje , title=. The Review of Financial Studies , year=. doi:10.1093/rfs/hhn098 , url=

    work page doi:10.1093/rfs/hhn098

  28. [28]

    Journal of Financial Econometrics , volume=

    Bucci, Andrea , title=. Journal of Financial Econometrics , volume=. 2020 , publisher=

    2020

  29. [29]

    Statistics and Computing , volume=

    Comparing unconstrained parametrization methods for return covariance matrix prediction , author=. Statistics and Computing , volume=. 2022 , publisher=

    2022

  30. [30]

    Computational Statistics , volume=

    Bucci, Andrea and Palomba, Giulio and Tedeschi, Marco , title=. Computational Statistics , volume=. 2026 , publisher=

    2026

  31. [31]

    2013 , publisher=

    Canova, Fabio and Ciccarelli, Matteo , title=. 2013 , publisher=

    2013

  32. [32]

    and Qiao, Zhuo and Wong, Wing-Keung , title =

    Chiang, Thomas C. and Qiao, Zhuo and Wong, Wing-Keung , title =. Journal of Forecasting , volume =. 2010 , pages =

    2010

  33. [33]

    Advances in neural information processing systems , volume=

    A statistical recurrent model on the manifold of symmetric positive definite matrices , author=. Advances in neural information processing systems , volume=

  34. [34]

    Callot, Laurent A. F. and Kock, Anders B. and Medeiros, Marcelo C. , title=. Journal of Applied Econometrics , year=. doi:10.1002/jae.2512 , url=

    work page doi:10.1002/jae.2512

  35. [35]

    and Hentschel, Ludger , title=

    Campbell, John Y. and Hentschel, Ludger , title=. Journal of Financial Economics , year=

  36. [36]

    The Journal of Finance , year =

    Chae, Joon , title =. The Journal of Finance , year =

  37. [37]

    and Gray, Philip , title=

    Chan, Kam F. and Gray, Philip , title=. Journal of Futures Markets , year=. doi:10.1002/fut.21922 , url=

    work page doi:10.1002/fut.21922

  38. [38]

    Journal of Econometrics , year=

    Chen, Rong and Xiao, Han and Yang, Dan , title=. Journal of Econometrics , year=

  39. [39]

    , title =

    Christie, Andrew A. , title =. Journal of Financial Economics , year =

  40. [40]

    Journal of Applied Econometrics , year =

    Halbleib-Chiriac, Roxana and Voev, Valeri , title =. Journal of Applied Econometrics , year =

  41. [41]

    Journal of Financial Economics , year =

    Chordia, Tarun and Roll, Richard and Subrahmanyam, Avanidhar , title =. Journal of Financial Economics , year =

  42. [42]

    , title=

    Copeland, Thomas E. , title=. The Journal of Finance , year=. doi:10.2307/2326280 , url=

    work page doi:10.2307/2326280

  43. [43]

    Journal of Financial Econometrics , year=

    Corsi, Fulvio , title=. Journal of Financial Econometrics , year=

  44. [44]

    Measuring the liquidity part of volume , journal=

    Darolles, Serge and Le Fol, Ga. Measuring the liquidity part of volume , journal=. 2015 , volume=

    2015

  45. [45]

    Mixture of distribution hypothesis: Analyzing daily liquidity frictions and information flows , journal=

    Darolles, Serge and Le Fol, Ga. Mixture of distribution hypothesis: Analyzing daily liquidity frictions and information flows , journal=. 2017 , volume=

    2017

  46. [46]

    The Review of Financial Studies , year=

    DeMiguel, Victor and Garlappi, Lorenzo and Uppal, Raman , title=. The Review of Financial Studies , year=. doi:10.1093/rfs/hhm075 , url=

    work page doi:10.1093/rfs/hhm075

  47. [47]

    Econometric Reviews , year =

    De Pooter, Michiel and Martens, Martin and Van Dijk, Dick , title =. Econometric Reviews , year =

  48. [48]

    and Yilmaz, Kamil , title=

    Diebold, Francis X. and Yilmaz, Kamil , title=. The Economic Journal , year=

  49. [49]

    Journal of Statistical Computation and Simulation , year=

    Dutilleul , Pierre , title=. Journal of Statistical Computation and Simulation , year=. doi:10.1080/00949659908811970 , url=

    work page doi:10.1080/00949659908811970

  50. [50]

    and Koloydenko, Alexey and Zhou, Diwei , title=

    Dryden, Ian L. and Koloydenko, Alexey and Zhou, Diwei , title=. The Annals of Applied Statistics , year=

  51. [51]

    and Mardia, Kanti V

    Dryden, Ian L. and Mardia, Kanti V. , publisher =. Procrustes analysis , booktitle =. doi:https://doi.org/10.1002/9781119072492.ch7 , url =

    work page doi:10.1002/9781119072492.ch7

  52. [52]

    , title=

    Engle, Robert F. , title=. Econometrica , year=

  53. [53]

    , title =

    Engle, Robert F. , title =. Journal of Business and Economic Statistics , year =

  54. [54]

    and Kroner, Kenneth Ff , title =

    Engle, Robert F. and Kroner, Kenneth Ff , title =. Econometric Theory , year =

  55. [55]

    The Journal of Finance , year=

    Eraker, Bj. The Journal of Finance , year=

  56. [56]

    , title=

    Fama, Eugene F. , title=. The Journal of Finance , year=

  57. [57]

    and Schwert, G.William and Stambaugh, Robert F

    French, Kenneth R. and Schwert, G.William and Stambaugh, Robert F. , title=. Journal of Financial Economics , year=

  58. [58]

    Ronald and Rossi, Peter E

    Gallant, A. Ronald and Rossi, Peter E. and Tauchen, George , title =. The Review of Financial Studies , year =

  59. [59]

    Quantitative Finance , year=

    Gatheral , Jim and Jaisson , Thibault and Rosenbaum , Mathieu , title=. Quantitative Finance , year=

  60. [60]

    Journal of Empirical Finance , year =

    Giot, Pierre and Laurent, S. Journal of Empirical Finance , year =

  61. [61]

    , title=

    Gower, J.C. , title=. Psychometrika , year=

  62. [62]

    Journal of Empirical Finance , year=

    Gribisch, Bastian and Hartkopf, Jan Patrick and Liesenfeld, Roman , title=. Journal of Empirical Finance , year=

  63. [63]

    Hall, Peter , title =

  64. [64]

    and Lunde, Asger and Nason, James M

    Hansen, Peter R. and Lunde, Asger and Nason, James M. , title =. Econometrica , year =

  65. [65]

    , title=

    Han, Chulwoo and Park, Frank C. , title=. Journal of Banking. 2022 , volume=

    2022

  66. [66]

    Computer Vision--ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part II 13 , pages=

    From manifold to manifold: Geometry-aware dimensionality reduction for SPD matrices , author=. Computer Vision--ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part II 13 , pages=. 2014 , organization=

    2014

  67. [67]

    The Journal of Financial and Quantitative Analysis , volume =

    Harris, Lawrence , title =. The Journal of Financial and Quantitative Analysis , volume =

  68. [68]

    , title=

    Hasbrouck, Joel and Seppi, Duane J. , title=. Journal of Financial Economics , year=

  69. [69]

    The Journal of Financial and Quantitative Analysis , year=

    He, Xiaojun and Velu, Raja , title=. The Journal of Financial and Quantitative Analysis , year=

  70. [70]

    Journal of Applied Econometrics , year=

    Herwartz, Helmut and Wang, Shu , title=. Journal of Applied Econometrics , year=

  71. [71]

    , title=

    Higham, Nicholas J. , title=. Linear Algebra and its Applications , year=

  72. [72]

    Huang, Zhiwu and Van Gool, Luc , booktitle=

  73. [73]

    and Moulines, Eric and Soulier, Philippe , title=

    Hurvich, Clifford M. and Moulines, Eric and Soulier, Philippe , title=. Econometrica , year=

  74. [74]

    , title=

    Jin, Xin and Maheu, John M. , title=. Journal of Econometrics , year=

  75. [75]

    The Journal of Finance , year=

    Johannes, Michael , title=. The Journal of Finance , year=. doi:10.1111/j.1540-6321.2004.00632.x , url=

    work page doi:10.1111/j.1540-6321.2004.00632.x 2004

  76. [76]

    , title =

    Karpoff, Jonathan M. , title =. The Journal of Financial and Quantitative Analysis , year =

  77. [77]

    The Review of Economics and Statistics , year=

    Kilian, Lutz , title=. The Review of Economics and Statistics , year=

  78. [78]

    Oxford Bulletin of Economics and Statistics , volume =

    Koop, Gary and Korobilis, Dimitris , title =. Oxford Bulletin of Economics and Statistics , volume =

  79. [79]

    , title =

    Kyle, Albert S. , title =. Econometrica , year =

  80. [80]

    and Lastrapes, William D

    Lamoureux, Christopher G. and Lastrapes, William D. , title =. The Journal of Finance , year =

Showing first 80 references.