Paper: arXiv 2408.03137

Authors: Abdulnasser Hatemi-J

Abstract

Asymmetric causality tests are increasingly gaining popularity in different scientific fields. This approach corresponds better to reality since logical reasons behind asymmetric behavior exist and need to be considered in empirical investigations. Hatemi-J (2012) introduced the asymmetric causality tests via partial cumulative sums for positive and negative components of the variables operating within the vector autoregressive (VAR) model. However, since the residuals across the equations in the VAR model are not independent, the ordinary least squares method for estimating the parameters is not efficient. Additionally, asymmetric causality tests mean having different causal parameters (i.e., for positive or negative components), thus, it is crucial to assess not only if these causal parameters are individually statistically significant, but also if their difference is statistically significant. Consequently, tests of difference between estimated causal parameters should explicitly be conducted, which are neglected in the existing literature. The purpose of the current paper is to deal with these issues explicitly. An application is provided, and ten different hypotheses pertinent to the asymmetric causal interaction between two largest financial markets worldwide are efficiently tested within a multivariate setting.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 7.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper introduces advanced econometric theory with matrix algebra and multivariate GARCH, indicating high mathematical complexity. It includes a specific empirical application testing ten hypotheses on major financial markets, demonstrating backtest-ready methodology and implementation focus, thus achieving high empirical rigor.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Efficient Asymmetric Causality Testing"] --> B["Data Input & Model Setup<br>Multivariate VAR with Partial Cumulative Sums"]
  B --> C["Computational Process<br>Estimate System with Feasible GLS for Efficiency"]
  C --> D["Statistical Testing Phase<br>Test Individual Parameters & Differences"]
  D --> E{"Key Findings & Outcomes<br>Ten Hypotheses Tested on Financial Markets"}

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper presents a system of equations that allows efficient estimation and testing of differences between causal parameters for positive and negative components, including multivariate GARCH and t‑distribution effects, and applies it to ten asymmetric causality hypotheses between the two largest

Source sentence
The current paper deals with these issues and present a system of equations for conducting the tests.

Assumes 4

  • The residuals across the equations in the VAR model are not independent, making ordinary least squares inefficient. model
    Source sentence
    However, since the residuals across the equations in the VAR model are not independent, the ordinary least squares method for estimating the parameters is not efficient. The critique of this approach is that the positive and negative components are not independent and therefore the ordinary least squares method used for estimating the parameters in the VAR model is not efficient in this case.
  • Positive and negative components of the variables are not independent. model
    Source sentence
    The critique of this approach is that the positive and negative components are not independent and therefore the ordinary least squares method used for estimating the parameters in the VAR model is not efficient in this case.
  • If the error terms exhibit clustering, the model can be estimated with multivariate GARCH effects combined with a multivariate t‑distribution to account for fat tails. statistical
    Source sentence
    The system can be estimated efficiently, and it can also account for the multivariate GARCH effects combined with the multivariate t-distribution accounting for potential fat-tails effects.
  • The variables have a unit root, requiring an extra lag of each variable in each equation. data
    Source sentence
    Since the variables have a unit root, an extra lag of each variable needs to be added into each equation in order to account for the impact of the unit root on causality tests according to Toda and Yamamoto (1995).5 3.

Data

  • US and Chinese share price indexes · FRED database market-prices public
    March 1999 to May 2024 · two largest financial markets
    Source sentence
    An Application The suggested asymmetric tests are applied to investigate the causal interaction between falling and rising prices in the two largest financial markets in the world. The monthly data for all share price indexes for the US and the Chinese markets are used for this purpose.6 The period covers March of 1999 until May of 2024.

Recognised sources: FRED

Costs

Not applicable The paper reports no trading or execution result.

Finds 4

  • The two financial markets are causally related and the causal impact of each market on the other is asymmetric. statistical
    Source sentence
    The results show that these two financial markets are causally related and the causal impact of each market on the other is asymmetric.
  • These asymmetric causal impacts are statistically significant. statistical
    Source sentence
    These asymmetric causal impacts between the markets are statistically significant.
  • The impact of falling Chinese markets on falling US markets is greater than the reverse impact. statistical
    Source sentence
    Furthermore, the estimations uncover that the impact of falling markets in China on the falling US markets is greater compared to the reverse impact.
  • Rising US markets have a stronger causal impact on rising Chinese markets compared to the reverse effect. statistical
    Source sentence
    However, the rising markets in the US have a stronger causal impact on the rising Chinese markets contrasted with the reverse causal effect.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 19 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (30)

  1. A bibliometric review of financial market integration literature (2022) — link
  2. Dynamic Asymmetric Causality Tests with an Application (2021) — link
  3. Theory I: Prediction (2021) — link
  4. ASYM_CAUS: C++ module for Transforming an Integrated Variable with Deterministic Trend Parts into Negative and Positive Cumulative Partial Sums (2017)
  5. A MS-Excel Module to Transform an Integrated Variable into Cumulative Partial Sums for Negative and Positive Components with and without Deterministic Trend Parts. (2016)
  6. TDICPS: OCTAVE module to Transform an Integrated Variable into Cumulative Partial Sums for Negative and Positive Components with Deterministic Trend Parts (2016)
  7. An extension of the asymmetric causality tests for dealing with deterministic trend components (2016) — link
  8. Testing for Multiple Bubbles: Historical Episodes of Exuberance and Collapse in the S&P 500 (2015) — link
and 22 more
  1. ASCOMP: GAUSS module to Transform Data into Cumulative Positive and Negative Components (2014)
  2. Asymmetric causality tests with an application (2012) — link
  3. Testing for Granger Non-causality in Heterogeneous Panels (2012) — link
  4. A bootstrap test for causality with endogenous lag length choice: theory and application in finance (2012) — link
  5. Asymmetric Panel Causality Tests with an Application to the Impact of Fiscal Policy on Economic Performance in Scandinavia (2011)
  6. Testing for Granger causality in heterogeneous mixed panels (2011) — link
  7. Multivariate GARCH models (2008) — link
  8. Exports and growth: Granger causality analysis on OECD countries with a panel data approach (2006) — link
  9. Tests for causality between integrated variables using asymptotic and bootstrap distributions: theory and application (2006) — link
  10. A test for multivariate ARCH effects (2005) — link
  11. Statistical inference in vector autoregressions with possibly integrated processes (1995) — link
  12. Causality, cointegration, and control (1988) — link
  13. Co-integration and error correction: representation, estimation and testing (1987) — link
  14. Measurement of Linear Dependence and Feedback between Multiple Time Series (1982) — link
  15. Distribution of the Estimators for Autoregressive Time Series with a Unit Root (1979) — link
  16. Spurious regressions in econometrics (1974) — link
  17. Investigating causal relations by econometric models and cross-spectral methods (1969) — link
  18. Statistical Decision Functions (1951) — link
  19. Asymmetric Generalized Impulse Response Functions with an Application in Finance (2014)
  20. HIDDEN COINTEGRATION (2002) — link
  21. Developments in the study of cointegrated economic variables (2001) — link
  22. Money, Income, and Causality (1972)

0 of 30 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.