Paper: arXiv 2408.12210

Authors: Cameron Cornell, Lewis Mitchell, Matthew Roughan

Abstract

Financial networks can be constructed using statistical dependencies found within the price series of speculative assets. Across the various methods used to infer these networks, there is a general reliance on predictive modelling to capture cross-correlation effects. These methods usually model the flow of mean-response information, or the propagation of volatility and risk within the market. Such techniques, though insightful, don’t fully capture the broader distribution-level causality that is possible within speculative markets. This paper introduces a novel approach, combining quantile regression with a piecewise linear embedding scheme - allowing us to construct causality networks that identify the complex tail interactions inherent to financial markets. Applying this method to 260 cryptocurrency return series, we uncover significant tail-tail causal effects and substantial causal asymmetry. We identify a propensity for coins to be self-influencing, with comparatively sparse cross variable effects. Assessing all link types in conjunction, Bitcoin stands out as the primary influencer - a nuance that is missed in conventional linear mean-response analyses. Our findings introduce a comprehensive framework for modelling distributional causality, paving the way towards more holistic representations of causality in financial markets.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 6.0/10 (capped from 7.0: no out-of-sample test; see penalty flags)
  • Quadrant: Holy Grail — high math complexity, high empirical rigor (effective empirical rigor capped from 7.0)

Why this score: The paper employs advanced statistical modeling techniques including piecewise linear embeddings and quantile regression, involving significant mathematical formulation; the methodology is applied to a large-scale, real-world dataset (260 cryptocurrencies) with detailed empirical results, though specific backtesting code or robustness checks are not explicitly shown in the excerpt.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Detecting distribution-level causality<br>beyond mean-response models"] --> B["Methodology<br>Quantile Regression + Piecewise Linear<br>Embedding for Tail Interactions"]
  B --> C["Data Input<br>260 Cryptocurrency<br>Return Series"]
  C --> D["Computational Process<br>Inference of Cross-Quantile<br>Causality Matrices"]
  D --> E{"Analysis of<br>Network Topology"}
  E --> F1["Key Finding 1<br>Significant Tail-Tail<br>Causal Effects"]
  E --> F2["Key Finding 2<br>Strong Causal<br>Asymmetry"]
  E --> F3["Key Finding 3<br>Bitcoin as Primary<br>Network Influencer"]

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 proposes a novel method that combines quantile regression with a piecewise linear embedding scheme to build distributional causality networks.

Source sentence
This paper introduces a novel approach, combining quantile regression with a piecewise linear embedding scheme — allowing us to construct causality networks that identify the complex tail interactions inherent to financial markets.

Assumes 4

  • Causal links are assumed to first appear in order‑1 VAR effects; higher‑order lags are not considered independent. model
    Source sentence
    For the empirical networks in Section 4 we operate under the assumption that any causal link i →j will first manifest in order-1 effects, and that lag p > 1 effects will not occur independently of a p = 1 dependence.
  • The VAR process is limited to order‑1 (VAR(1)) for analysis. model
    Source sentence
    To streamline our discussion and estimation of causal networks, we limit our analysis to VAR(1) processes and omit the index p from our discussion, with A = A1 unless otherwise specified.
  • Error terms are assumed to have zero mean; Gaussianity is not required. statistical
    Source sentence
    The vector ϵt is often assumed to be Gaussian, however this is not a strict requirement.
  • Hourly returns are transformed by sign(yt) × log(|yt| + 1) to mitigate data artifacts. data
    Source sentence
    To minimise the effects of data artifacts, such as coordinated pump and dump schemes, we take a stabilising transformation sign(yt) × log(|yt| + 1).

Data

  • Hourly cryptocurrency price series market-prices unknown
    2021-01-01 to 2022-01-01 · 260 cryptocurrencies selected from top 750 by market cap
    Source sentence
    Our dataset comprises hourly prices for 260 cryptocurrencies spanning from January 1, 2021, to January 1, 2022, along with their market capitalizations as of June 2022.

Costs

Not applicable The paper reports no trading or execution result.

Finds 4

  • Bitcoin emerges as the primary influencer in the full causality network, a nuance missed by conventional mean‑response analyses. statistical
    Source sentence
    Assessing all link types in conjunction, Bitcoin stands out as the primary influencer - a nuance that is missed in conventional linear mean-response analyses.
  • Self‑causal links are highly prevalent, with 65.8% of possible self‑links statistically significant, showing strong mean‑reversion and tail‑tail excitation. statistical
    Source sentence
    In general, there is a high proportion of potential self-links that are statistically significant (65.8% of possible links), exhibiting pronounced directional trends.
  • Upper‑Upper tail link is the strongest effect, present in 95% of possible links and all positively signed. statistical
    Source sentence
    The Upper-Upper link is the strongest of all effect, with 95% of possible links being present — all of which are positively signed.
  • Cross‑causal effects are sparse, with only 4.5% of possible off‑diagonal links statistically significant and relatively symmetric. statistical
    Source sentence
    Generally, there is a greatly reduced number of statistically significant effects (4.5% of possible links), with a comparatively high degree of symmetry.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 16 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 1 paper in the archive

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 (66)

  1. Macroeconomics and Reality (1980)
  2. Measures of Causality in Complex Datasets with Application to Financial Data (2014)
  3. Networks: an introduction (2010)
  4. Bayesian Graphical Models for Structural Vector Autoregressive Processes (2016)
  5. Clustering in Complex Directed Networks (2007)
  6. Cryptocurrency market structure: connecting emotions and economics (2019)
  7. Predicting future stock market structure by combining social and financial network information (2019)
  8. Dominating Clasp of the Financial Sector Revealed by Partial Correlation Analysis of the Stock Market (2010)
and 58 more
  1. Statistical analysis of financial networks (2005)
  2. Clustering and information in correlation based financial networks (2003)
  3. Structural Entropy: Monitoring Correlation-Based Networks Over Time With Application To Financial Markets (2019)
  4. Information Network Modeling for U.S. Banking Systemic Risk (2020)
  5. Econometric Measures of Connectedness and Systemic Risk in the Finance and Insurance Sectors (2011)
  6. On the Network Topology of Variance Decompositions: Measuring the Connectedness of Financial Firms (2011)
  7. NETS: Network Estimation for Time Series (2018)
  8. Bayesian Graphical Models for Structural Vector Autoregressive Processes (2012)
  9. Network based evidence of the financial impact of Covid -19 pandemic (2021)
  10. An Analysis of Variance Test for Normality (Complete Samples) (1965)
  11. Causation Entropy Identifies Indirect Influences, Dominance of Neighbors and Anticipatory Couplings (2014)
  12. Cryptocurrencies: An unconventional challenge to the AML/CFT regulators? (2014)
  13. Vector autoregression and causality: a theoretical overview and simulation study (1994)
  14. Vector Autoregression and Causality: A Theoretical Overview and Simulation Study (1991)
  15. Emergence of Scaling in Random Networks (1999)
  16. Granger Causality and Transfer Entropy Are Equivalent for Gaussian Variables (2009) — link
  17. The New Introduction to Multiple Time Series Analysis (2005)
  18. The Advanced Theory of Statistics, Volume 2: Inference and Relationship (1973)
  19. Cryptocurrencies, Mainstream Asset Classes and Risk Factors - A Study of Connectedness (2018)
  20. Modelling Determinants of Cryptocurrency Prices: A Bayesian Network Approach (2023) — link
  21. Causality and dynamic spillovers among cryptocurrencies and currency markets (2020)
  22. Causal inference between cryptocurrency narratives and prices: Evidence from a complex dynamic ecosystem (2020)
  23. NETS: Network estimation for time series (2019) — link
  24. Econometric Measures of Connectedness and Systemic Risk in the Finance and Insurance Sectors (2011)
  25. A representation theory for a class of vector autoregressive models for fractional processes (2008)
  26. Robust Estimation of the Vector Autoregressive Model by a Least Trimmed Squares Procedure (2008)
  27. Robust estimation for vector autoregressive models (2013)
  28. A discussion on the robust vector autoregressive models: novel evidence from safe haven assets (2022)
  29. Cryptocurrencies: market analysis and perspectives (2020)
  30. Identifying and Analyzing Cryptocurrency Manipulations in Social Media (2021)
  31. How Elon Musk’s Twitter activity moves cryptocurrency markets (2023) — link
  32. Vector Autoregression in Cryptocurrency Markets: Unraveling Complex Causal Networks — → 2308.15769
  33. The Use of the Rank Transform in Regression (1979)
  34. Rank regression: an alternative regression approach for data with outliers (2014)
  35. The impact of COVID-19 on the degree of dependence and structure of risk-return relationship: A quantile regression approach (2020) — link
  36. Does implied volatility (or fear index) affect Islamic stock returns and conventional stock returns differently? Wavelet-based granger-causality, asymmetric quantile regression and NARDL approaches (2022) — link
  37. CAViaR: Conditional Autoregressive Value at Risk by Regression Quantiles (2004) — link
  38. Robust Tests for Heteroscedasticity Based on Regression Quantiles (1982)
  39. Regression Quantiles (1978)
  40. Calculation of Quantile Regression Standard Errors (1993)
  41. Econometric Analysis (2012)
  42. On the Distribution of a Studentized Quantile (1988)
  43. Efficiency, multifractality, and the long-memory property of the Bitcoin market: A comparative analysis with stock, currency, and gold markets (2018) — link
  44. The structure and degree of dependence: A quantile regression approach (2013) — link
  45. Causality in quantiles and dynamic stock return–volume relations (2009) — link
  46. Does uncertainty move the gold price? New evidence from a nonparametric causality-in-quantiles test (2016) — link
  47. Distributional predictability between commodity spot and futures: Evidence from nonparametric causality-in-quantiles tests (2019) — link
  48. Asymmetric dependence between economic policy uncertainty and stock market returns in G7 and BRIC: A quantile regression approach (2018) — link
  49. Quantile Autoregression (2006) — link
  50. Network quantile autoregression (2019) — link
  51. Asymmetric volatility in cryptocurrency markets: New evidence from smooth transition GARCH models (2020) — link
  52. Volatility dynamics of crypto-currencies’ returns: Evidence from asymmetric and long memory GARCH models (2020) — link
  53. GARCH Modelling of Cryptocurrencies (2017) — link
  54. Performance of ARCH and GARCH Models in Forecasting Cryptocurrency Market Volatility (2021) — link
  55. Statsmodels: Econometric and statistical modeling with python (2010)
  56. Network vector autoregression (2017) — link
  57. Rank is all you need: development and analysis of robust causal networks (2024) — link
  58. Rank Is All You Need: Robust Estimation of Complex Causal Networks (2024)

1 of 66 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.