Pith. sign in

REVIEW 4 major objections 7 minor 1 cited by

Sequential Monte Carlo for Noncausal Processes

T0 review · 4 major / 7 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A sequential Monte Carlo sampler estimates noncausal vector autoregressions and jointly selects orders and error distribution.

desk verdict A useful but uneven SMC adaption for noncausal VARs: the estimator works, but the MDD identification leg is weaker than advertised and the printed MDD formula needs fixing. read the letter →

arxiv 2501.03945 v1 pith:2FILVE37 submitted 2025-01-07 econ.EM

Gianluca Cubadda , Francesco Giancaterini , Stefano Grassi This is my paper · ORCID
classification econ.EM MSC 62F1562M1065C05
keywords mixedcausal-noncausalautoregressivemodelssequentialMonteCarloBayesianestimationmarginaldatadensitymodelselectionheavy-tailederrordistributionsvectorautoregressionnoncausalidentification
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper sets out to show that Sequential Monte Carlo, an algorithm that evolves a swarm of parameter draws through gradually more data-informed distributions, is a practical and fast Bayesian estimator for mixed causal-noncausal autoregressive models. The payoff is that the same sampler works for Cauchy, Student-t, and Skewed-t errors and returns the marginal data density as a byproduct, so model selection can choose the causal and noncausal order and the error distribution in one pass. Simulation evidence shows accurate posterior means and mostly correct identification; the authors report that BIC identifies the true model more often than MDD when the error is Skewed-t. This matters because noncausal models are used to model bubbles and expectation-driven dynamics, and existing Bayesian tools are tied to Student-t errors and are slower to run.

What carries the argument

The load-bearing object is the tempered Sequential Monte Carlo sampler: a swarm of $P$ particles is drawn from the prior, then moved through $M$ bridge distributions $\pi_m(\Theta)\propto [p(y|\Theta)]^{\rho_m}p(\Theta)$ with $\rho_m=((m-1)/(M-1))^\lambda$, using correction (reweighting), selection (resampling when the effective sample size drops below $P/2$), and mutation (short MCMC steps with invariant distribution $\pi_m$). This schedule is what lets the sampler find multiple modes instead of getting stuck in local minima, and it makes the marginal data density $\hat p(y)=\prod_{m=1}^M \sum_i \tilde w^i_m$ a free byproduct rather than a separate computation. The supporting identification machinery is the grid of 21 candidate models (seven order pairs across three error distributions), compared by MDD and BIC.

What would settle it

Rerun the Skewed-t VMAR(1,1) Monte Carlo with larger settings such as $M=300$ and $P=30000$: if the MDD identification rate stays near 62% while BIC stays near 94%, the fixed tuning is the bottleneck and the MDD-based identification claim is conditional on it. A direct check is to compare the SMC marginal data density against a high-accuracy estimate from a much longer MCMC run; systematic disagreement for Skewed-t data would show the byproduct MDD is not reliable enough for model choice.

Watch

Extended reading notes

Core claim

The paper claims that the Sequential Monte Carlo sampler of Herbst and Schorfheide (2014) and Bognanni and Herbst (2018) can serve as a general Bayesian engine for vector mixed causal-noncausal autoregressions $\Psi(L)\Phi(L^{-1})y_t=u_t$, whatever the error distribution, as long as the posterior kernel can be evaluated pointwise. It supplies posterior mean estimates and, without extra computation, the marginal data density, which together with BIC selects the causal order $r$, noncausal order $s$, and the error law. In Monte Carlo exercises with Cauchy, Student-t, and Skewed-t errors, the estimates have small bias and RMSE, and both criteria usually recover the true VMAR(1,1); the paper reports that BIC outperforms MDD when the errors are Skewed-t. Applied to detrended monthly S&P Europe 350 ESG Index and Brent oil prices, the procedure chooses a bivariate VMAR(1,1) with Student-t errors and finds positive cross-effects in both directions, which the authors read as expectation-driven dynamics.

Load-bearing premise

The central assumption is that the SMC estimate of the marginal data density is accurate enough at the fixed settings $P=10000$, $M=100$, $\lambda=2$ to drive model selection; if that estimate is biased for heavy-tailed or asymmetric posteriors, the proposed Bayes-factor identification fails even when posterior means are close to the truth.

Editorial extensions

If this is right

  • For any error distribution with a pointwise-evaluable posterior kernel, the same SMC code replaces bespoke MCMC samplers, removing the Student-t-only restriction in Bayesian noncausal modelling.
  • Marginal-data-density computation becomes free, so Bayes factors for comparing $r,s$ and error laws no longer require a separate marginal-likelihood step.
  • Because particles are independent, estimation can be parallelized across cores, cutting wall-clock time that currently limits Monte Carlo studies and empirical applications.
  • In the ESG/oil application, the selected VMAR(1,1) with Student-t errors implies that expected ESG index movements feed into Brent prices and vice versa, giving a testable forward-looking channel.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The underperformance of MDD under Skewed-t at the chosen tuning suggests that users should either prefer BIC for heavy-tailed asymmetric DGPs or increase $M$ and $P$; the paper itself notes more stages would help but become computationally costly.
  • The same sampler could be extended to other non-Gaussian error families, Markov-switching parameters, or stochastic volatility, since the only requirement is pointwise kernel evaluation.
  • A practical test of the identification rule would be to compare SMC MDD values against a long-run MCMC marginal-likelihood estimate; if they disagree mainly for Skewed-t data, a correction or adaptive tempering would be needed.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 7 minor

Summary. The paper proposes a Sequential Monte Carlo (SMC) algorithm for Bayesian estimation of mixed causal-noncausal vector autoregressive models (VMAR(r,s)) with Student-t, Cauchy, and Skewed-t error distributions. It introduces a model-selection procedure, based on the SMC-computed marginal data density (MDD) and the Bayesian information criterion (BIC), that chooses both the causal and noncausal polynomial orders and the innovation distribution. Monte Carlo experiments with T=150 and B=200 replications for a bivariate VMAR(1,1) report posterior estimates close to the true values under all three error distributions, with BIC selecting the true specification in 83--94% of replications. The MDD selection is strong for Cauchy and Student-t DGPs but drops to 62% for the Skewed-t DGP. An empirical application to the S&P Europe 350 ESG Index and Brent crude oil prices selects a VMAR(1,1) model with Student-t errors.

Significance. If the numerical issues identified below are corrected, the paper would offer a useful computational alternative for noncausal models: SMC avoids the mode search and MCMC tuning of existing methods, is parallelizable, computes the MDD as a byproduct, and accommodates heavy-tailed and skewed innovation distributions. The simultaneous identification of polynomial orders and error distribution is a genuine extension of previous Bayesian noncausal VAR work. The Monte Carlo point estimates in Table 2 are mostly close to their true values, and BIC identifies the correct specification in a large majority of replications. However, the current manuscript contains a likelihood formula error, an unnormalized MDD formula, and a strong MDD-selection failure for the Skewed-t DGP, so the abstract's claim of 'strong ability ... to correctly identify the process' is not yet supported.

major comments (4)
  1. [Section 4.2, Eqs. (9)-(11)] The likelihood is written as a sum over t of the innovation densities, and Eq. (9) repeats the sum inside the density. For an i.i.d. innovation sequence the likelihood is the product over t of the densities, not the sum. As printed, the posterior is not the posterior of the stated model; if the implementation actually uses a product, the displayed formulas should be corrected. This is load-bearing because both MDD and BIC are computed from this likelihood.
  2. [Section 3.2, Eq. (5), and Appendix B] The MDD estimator is printed as the product over m of the unnormalized sum of incremental weights, with no 1/P factor, while the correction step in Appendix B normalizes the incremental weights by (1/P) times the sum. The correct estimator is the product over m of (1/P) times the sum of the incremental weights. The missing factor changes the numerical MDD by P^M, so the MDD levels reported in Table 3 are not valid as printed. The factor cancels in comparisons that use the same P, but the formula must be corrected and independently validated.
  3. [Section 4.3, Table 1] Under the Skewed-t DGP, the MDD criterion selects the true VMAR(1,1) with Skewed-t errors only 62.0% of the time and instead selects a Student-t error distribution for the same orders 21.5% of the time, while BIC achieves 94.0%. Since P=10000, M=100, and lambda=2 are the fixed tuning values used throughout the Monte Carlo study, the abstract's 'strong ability' claim is not supported for one of the three error distributions considered. The paper should either temper the claim, demonstrate a practically feasible configuration that fixes the MDD performance, or benchmark the MDD estimator against an independent marginal-likelihood estimator.
  4. [Table 2, Skewed-t DGP rows for alpha_1 and alpha_2] The reported average of the per-replication standard errors (STD) is 0.0117 and 0.0143 for alpha_1 and alpha_2, respectively, while the Monte Carlo variance of the point estimates is 0.2349 and 0.2931, corresponding to a Monte Carlo standard deviation of about 0.48, and the RMSE is 0.4950 in both cases. The per-replication standard errors are implausibly small relative to the dispersion of the point estimates, suggesting an error in the computation or reporting of posterior standard deviations. This matters because the accuracy claim for the skewness parameters rests on these numbers.
minor comments (7)
  1. [Table 1 caption] The caption says 'MMD' where the text and table body use 'MDD'; this should be corrected.
  2. [Table 3, BIC row] The BIC value for the Student-t VMAR(0,1) model is reported as 1521.6, identical to the Cauchy VMAR(0,1) value; this is likely a typo and should be checked.
  3. [Section 3.2, Eq. (5)] The expression uses the symbol n for the stage index in the incremental weights, but the stage index elsewhere is m; use a consistent notation.
  4. [Section 3.1] The definition of Theta_1 lists vec(Psi_1),...,vec(Psi_s) and vec(Phi_1),...,vec(Phi_r), but Psi is the causal polynomial of order r and Phi is the noncausal polynomial of order s; the indices appear to be swapped and should be corrected.
  5. [Section 4.2, likelihood definition] The notation p(y|Theta) = sum over t of p(u_t|Theta) appears before Eq. (9); if the intended likelihood is a product, the summation notation should be removed throughout Sections 4.2 and Appendix A.
  6. [Section 5 and Figure 1] The data are described as monthly from July 2017 to February 2024, but the horizontal axis of Figure 1 starts in 2015; please clarify the discrepancy.
  7. [Reproducibility] The paper does not report random seeds or provide replication code; adding code or a detailed pseudocode with tuning choices would strengthen reproducibility.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the SMC estimation and MDD/BIC identification are standard Bayesian computations whose outputs are not baked into their inputs; self-citations are motivational and non-load-bearing.

full rationale

The paper's derivation chain is self-contained. The SMC algorithm follows Herbst and Schorfheide (2014) and Bognanni and Herbst (2018); the likelihoods and priors are standard VMAR(r,s) objects from Lanne and Saikkonen (2013) and Lanne and Luoto (2016). The MDD in Eq. (5) and the incremental weights in Appendix B are the very quantities defined by the SMC literature, and the model selection exercise compares models fitted on the same data, which is standard Bayesian model comparison rather than a circular reduction. Table 1's weaker MDD performance under the Skewed-t DGP is a robustness/correctness concern, not a circular one; the paper itself acknowledges in Section 4.3 that more stages may be required and that increasing M is computationally impractical. The missing 1/P normalization in Eq. (5), if taken literally, changes MDD levels but cancels across model comparisons using the same P, and in any event is a computational error rather than a circular step. Self-citations (Cubadda et al. 2024 for local minima in the GCov objective function; Cubadda et al. 2023 for the non-Gaussian identification condition; Giancaterini et al. 2022 for the alpha=1 equivalence of Student-t and Skewed-t densities) are peripheral: they motivate the SMC choice or state known distributional identities, and none is used to define the target results of posterior estimation or MDD/BIC identification. No fitted parameter is relabeled as a prediction, and no quantity is defined in terms of the quantity it is said to select. The central claims may have accuracy limitations, and a direct validation of the MDD estimator against an independent marginal-likelihood estimator would strengthen the identification claim, but these are not circularity.

Assumptions & free parameters 9 free parameters · 7 assumptions · 0 invented entities

The central claim rests on standard Bayesian and SMC machinery plus domain assumptions specific to noncausal identification. No scientific entities are invented, but several tuning constants and prior hyperparameters are chosen by hand and not subjected to sensitivity analysis.

free parameters (9)
  • Tempering exponent lambda = 2
    Chosen in Section 4.3 and used again in Section 5; controls the rate at which likelihood information is introduced. No sensitivity analysis is provided.
  • Number of SMC stages M = 100 (simulations), 150 (empirical)
    Chosen by the authors; the paper notes increasing M is computationally impractical, yet M directly affects the quality of the MDD estimate.
  • Number of particles P = 10000 (simulations), 50000 (empirical)
    Chosen by the authors; particle count determines SMC approximation error and MDD bias.
  • MCMC mutation steps S
    Algorithm 1 says S iterations of MCMC are used but no value or adaptation rule is given.
  • Minnesota-type prior shrinkage gamma=2/i and delta=2/q = 2/i, 2/q
    Prior variances in Eq. (6), chosen to be more diffuse than Lanne and Luoto (2016). Posterior estimates depend on these choices.
  • Inverse Wishart prior hyperparameters Psi0=5I and nu~=3 = Psi0=5I, nu~=3
    Section 4.1 sets a diffuse prior for the scale matrix Sigma.
  • Exponential prior rate nu0=5 for degrees of freedom = 5
    Section 4.1 fixes the prior for the Student-t and Skewed-t degrees of freedom.
  • Skewness prior kappa=3 = 3
    Section 4.1 fixes the prior variance for the skewness parameter alpha.
  • Detrending polynomial order = 3
    Section 5 detrends the empirical series with a third-order polynomial; no sensitivity analysis is reported.
assumptions (7)
  • domain assumption Causal and noncausal polynomials have all roots outside the unit circle: psi(z) != 0 for |z| <= 1 and phi(z) != 0 for |z| <= 1.
    Invoked in Section 2.1 to define stationarity of MAR and VMAR processes; required for the likelihood and the two-sided moving average representation.
  • domain assumption The error term u_t is i.i.d. and non-Gaussian, which is necessary for identification of the noncausal component.
    Stated in Sections 2.1 and 2.2 following Breidt et al. (1991) and Lanne and Saikkonen. If errors were Gaussian, causal and noncausal factorizations would be observationally equivalent.
  • domain assumption The approximate likelihood p(y|Theta) = product over t=r+1 to T-s of p(u_t|Theta), which discards boundary terms, is a valid stand-in for the exact likelihood.
    Section 4.2 uses this likelihood for all SMC weights and MDD/BIC calculations. The approximation is inherited from Lanne and Saikkonen (2013), but no evidence is given for its accuracy at T=150 or for Cauchy and Skewed-t innovations.
  • ad hoc to paper The MDD estimator in Eq. (5), based on tempered incremental weights, provides a sufficiently accurate marginal likelihood for model comparison.
    The SMC cascade estimator is borrowed from Bognanni and Herbst (2018). Table 1 shows its selection rate under Skewed-t is only 62 percent, so accuracy at the chosen P, M, and lambda is not guaranteed.
  • domain assumption The candidate set of error distributions (Cauchy, Student-t, Skewed-t) and orders r,s in {0,1,2} contains the true data-generating process.
    Section 4.3 constructs DGPs from these distributions and estimates only these 21 models. The empirical application similarly assumes the true error distribution is among the three candidates.
  • standard math SMC importance-sampling convergence claims from Geweke (1989) and Bognanni and Herbst (2018) apply to the VMAR posterior kernels.
    Section 3.2 cites these results for consistency of the weighted Monte Carlo estimate and for the validity of the correction, selection, and mutation steps.
  • ad hoc to paper Detrending the empirical series with a third-order polynomial removes the stochastic trends while preserving bubble dynamics.
    Section 5 uses this detrending without sensitivity analysis; the maintained assumption is that the residual series is generated by a stationary VMAR(1,1) process.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Sequential Monte Carlo for Noncausal Processes." pith.science (2026). https://pith.science/paper/2FILVE37

@misc{pith2026250103945,
  author       = {Pith},
  title        = {Pith review of: Sequential Monte Carlo for Noncausal Processes},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2FILVE37}},
  note         = {Machine review of arXiv:2501.03945}
}
abstract

This paper proposes a Sequential Monte Carlo approach for the Bayesian estimation of mixed causal and noncausal models. Unlike previous Bayesian estimation methods developed for these models, Sequential Monte Carlo offers extensive parallelization opportunities, significantly reducing estimation time and mitigating the risk of becoming trapped in local minima, a common issue in noncausal processes. Simulation studies demonstrate the strong ability of the algorithm to produce accurate estimates and correctly identify the process. In particular, we propose a novel identification methodology that leverages the Marginal Data Density and the Bayesian Information Criterion. Unlike previous studies, this methodology determines not only the causal and noncausal polynomial orders but also the error term distribution that best fits the data. Finally, Sequential Monte Carlo is applied to a bivariate process containing S$\&$P Europe 350 ESG Index and Brent crude oil prices.

Figures

Figures reproduced from arXiv: 2501.03945 by the authors.

Figure 1
Figure 1. S&P Europe 350 Index and Brent prices 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 20 40 60 80 100 120 140 160 180 (a) The graph displays the S&P Europe 350 Index (solid black line) and Brent price (dashed line). The trend component is obtained using a third-order polynomial. 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 -40 -30 -20 -10 0 10 20 30 40 (b) The graph displays the detrended series of the S&P Eu… view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Regularized Generalized Covariance (RGCov) Estimator

    econ.EM 2025-04 conditional novelty 6.0 of 10

    A ridge-regularized Generalized Covariance estimator is proposed for high-dimensional mixed causal-noncausal VAR models, with asymptotic normality and chi-square tests when the shrinkage goes to zero.

Reference graph

Works this paper leans on

34 extracted references · 34 canonical work pages · cited by 1 Pith paper

  1. [1]

    Barigozzi, and M

    Alessi, L., M. Barigozzi, and M. Capasso (2011). Non-Fundamentalness in Structural Econometric Models: A Review . International Statistical Review\/ 79 , 16--47

    work page 2011

  2. [2]

    Azzalini, A. and A. Capitanio (2003). Distributions Generated by Perturbation of Symmetry With Emphasis on a Multivariate Skew-t Distribution . Journal of the Royal Statistical Society Series B: Statistical Methodology\/ 65 , 367--389

    work page 2003

  3. [3]

    Bec, F., H. B. Nielsen, and S. Saidi (2020). Mixed Causal Noncausal Autoregressions: Bimodality Issues in Estimation and Unit Root Testing 1 . Oxford Bulletin of Economics and Statistics\/ 82 , 1413--1428

    work page 2020

  4. [4]

    Bognanni, M. and E. Herbst (2018). A Sequential Monte Carlo Approach to Inference in Multiple-Equation Markov-Switching Models . Journal of Applied Econometrics\/ 33 , 126--140

    work page 2018

  5. [5]

    Breidt, F. J., R. A. Davis, K.-S. Lh, and M. Rosenblatt (1991). Maximum Likelihood Estimation for Noncausal Autoregressive Processes . Journal of Multivariate Analysis\/ 36 , 175--198

    work page 1991

  6. [6]

    Cavaliere, G., H. B. Nielsen, and A. Rahbek (2020). Bootstrapping Noncausal Autoregressions: With Applications to Explosive Bubble Modeling . Journal of Business & Economic Statistics\/ 38 , 55--67

    work page 2020

  7. [7]

    Giancaterini, A

    Cubadda, G., F. Giancaterini, A. Hecq, and J. Jasiak (2024). Optimization of the Generalized Covariance Estimator in Noncausal Processes . Statistics and Computing\/ 34 , 127

    work page 2024

  8. [8]

    Hecq, and E

    Cubadda, G., A. Hecq, and E. Voisin (2023). Detecting Common Bubbles in Multivariate Mixed Causal--Noncausal Models . Econometrics\/ 11 , 9

    work page 2023

Show all 34 references
  1. [9]

    Davis, R. A. and L. Song (2020). Noncausal Vector AR Processes with Application to Economic Time Series . Journal of Econometrics\/ 216 , 246--267

    2020

  2. [10]

    Durham, G. and J. Geweke (2014). Adaptive Sequential Posterior Simulators for Massively Parallel Computing Environments . In Bayesian model comparison , pp.\ 1--44. Emerald Group Publishing Limited

    2014

  3. [11]

    and J.-M

    Fries, S. and J.-M. Zakoian (2019). Mixed Causal-Noncausal AR Processes and the Modelling of Explosive Bubbles . Econometric Theory\/ 35 , 1234--1270

    2019

  4. [12]

    Geweke, J. (1989). Bayesian Inference in Econometric Models Using Monte Carlo Integration . Econometrica: Journal of the Econometric Society\/ , 1317--1339

    1989

  5. [13]

    Geweke, J. and M. Keane (2005). Bayesian Cross-Sectional Analysis of the Conditional Distribution of Earnings of Men in the United States, 1967-1996

    2005

  6. [14]

    Hecq, and C

    Giancaterini, F., A. Hecq, and C. Morana (2022). Is Climate Change Time-Reversible? Econometrics\/ 10 , 36

    2022

  7. [15]

    Gourieroux, C. and J. Jasiak (2017). Noncausal Vector Autoregressive Process: Representation, Identification and Semi-Parametric Estimation . Journal of Econometrics\/ 200 , 118--134

    2017

  8. [16]

    Gourieroux, C. and J. Jasiak (2018). Misspecification of Noncausal Order in Autoregressive Processes . Journal of Econometrics\/ 205 , 226--248

    2018

  9. [17]

    Gourieroux, C. and J. Jasiak (2023). Generalized Covariance Estimator . Journal of Business & Economic Statistics\/ 41 , 1315--1327

    2023

  10. [18]

    and J.-M

    Gourieroux, C. and J.-M. Zakoian (2015). On Uniqueness of Moving Average Representations of Heavy-Tailed Stationary Processes . Journal of Time Series Analysis\/ 36 , 876--887

    2015

  11. [19]

    and J.-M

    Gouri \'e roux, C. and J.-M. Zako \"i an (2017). Local Explosion Modelling by Non-causal Process . Journal of the Royal Statistical Society Series B: Statistical Methodology\/ 79 , 737--756

    2017

  12. [20]

    Zakoian, et al

    Gouri \'e roux, C., J.-M. Zakoian, et al. (2013). Explosive Bubble Modelling by Noncausal Process . CREST

    2013

  13. [21]

    Lieb, and S

    Hecq, A., L. Lieb, and S. Telg (2016). Identification of Mixed Causal-Noncausal Models in Finite Samples . Annals of Economics and Statistics/Annales d' \'E conomie et de Statistique\/ , 307--331

    2016

  14. [22]

    Hecq, A. and D. Velasquez-Gaviria (2022). Spectral Estimation for Mixed Causal-Noncausal Autoregressive Models . arXiv preprint arXiv:2211.13830\/

    2022 arXiv

  15. [23]

    Hecq, A. and E. Voisin (2021). Forecasting Bubbles with Mixed Causal-Noncausal Autoregressive Models . Econometrics and Statistics\/ 20 , 29--45

    2021

  16. [24]

    Hencic, A. and C. Gouri \'e roux (2015). Noncausal Autoregressive Model in Application to Bitcoin/USD Exchange Rates . Econometrics of risk\/ 583 , 17--40

    2015

  17. [25]

    Herbst, E. and F. Schorfheide (2014). Sequential Monte Carlo Sampling for DSGE Models . Journal of Applied Econometrics\/ 29 , 1073--1098

    2014

  18. [26]

    Luoma, and J

    Lanne, M., A. Luoma, and J. Luoto (2012). Bayesian Model Selection and Forecasting in Noncausal Autoregressive Models . Journal of Applied Econometrics\/ 27 , 812--830

    2012

  19. [27]

    Lanne, M. and J. Luoto (2016). Noncausal Bayesian Vector Autoregression . Journal of Applied Econometrics\/ 31 , 1392--1406

    2016

  20. [28]

    Lanne, M. and P. Saikkonen (2011). Noncausal Autoregressions for Economic Time Series . Journal of Time Series Econometrics\/ 3

    2011

  21. [29]

    Lanne, M. and P. Saikkonen (2013). Noncausal Vector Autoregression . Econometric Theory\/ 29 , 447--481

    2013

  22. [30]

    Moussa, Z. and A. Thomas (2023). Identifying Oil Supply News Shocks and Their Effects on the Global Oil Market . USAEE Working Paper\/

    2023

  23. [31]

    Nyberg, H. and P. Saikkonen (2014). Forecasting with a Noncausal VAR Model . Computational statistics & data analysis\/ 76 , 536--555

    2014

  24. [32]

    Phillips, P. C., S. Shi, and J. Yu (2015). Testing for Multiple Bubbles: Limit Theory of Real-Time Detectors . International Economic Review\/ 56 , 1079--1134

    2015

  25. [33]

    Phillips, P. C., Y. Wu, and J. Yu (2011). Explosive Behavior in the 1990s Nasdaq: When Did Exuberance Escalate Asset Values? International economic review\/ 52 , 201--226

    2011

  26. [34]

    Proietti, T. (2023). Peaks, Gaps, and Time-Reversibility of Economic Time Series . Journal of Time Series Analysis\/ 44 , 43--68

    2023

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.