Paper: arXiv 2308.15769
Authors: Cameron Cornell, Lewis Mitchell, Matthew Roughan
Abstract
Methodologies to infer financial networks from the price series of speculative assets vary, however, they generally involve bivariate or multivariate predictive modelling to reveal causal and correlational structures within the time series data. The required model complexity intimately relates to the underlying market efficiency, where one expects a highly developed and efficient market to display very few simple relationships in price data. This has spurred research into the applications of complex nonlinear models for developed markets. However, it remains unclear if simple models can provide meaningful and insightful descriptions of the dependency and interconnectedness of the rapidly developed cryptocurrency market. Here we show that multivariate linear models can create informative cryptocurrency networks that reflect economic intuition, and demonstrate the importance of high-influence nodes. The resulting network confirms that node degree, a measure of influence, is significantly correlated to the market capitalisation of each coin (\(ρ=0.193\)). However, there remains a proportion of nodes whose influence extends beyond what their market capitalisation would imply. We demonstrate that simple linear model structure reveals an inherent complexity associated with the interconnected nature of the data, supporting the use of multivariate modelling to prevent surrogate effects and achieve accurate causal representation. In a reductive experiment we show that most of the network structure is contained within a small portion of the network, consistent with the Pareto principle, whereby a fraction of the inputs generates a large proportion of the effects. Our results demonstrate that simple multivariate models provide nontrivial information about cryptocurrency market dynamics, and that these dynamics largely depend upon a few key high-influence coins.
Complexity vs Empirical Score
- Math Complexity: 4.5/10
- Empirical Rigor: 7.0/10
- Quadrant: Street Traders — empirical and applied, lower math complexity
Why this score: The paper employs Vector Autoregression (VAR) and statistical network analysis, which involves moderate mathematical complexity, but focuses heavily on empirical data application with specific datasets (261 cryptocurrencies, hourly returns) and statistical metrics (Spearman correlation, Gini coefficient), indicating a street-trader approach.
Research Flowchart
flowchart TD
A["Research Goal: Assess if simple multivariate models can describe cryptocurrency market dynamics?"] --> B["Methodology: Vector Autoregression & Causal Network Inference"]
B --> C["Data: Multivariate Price Series of Speculative Crypto Assets"]
C --> D["Computational Process: Model Complexity vs. Market Efficiency Analysis"]
D --> E{"Linear Model Application"}
E -->|"Success"| F["Outcome 1: Informative Networks reflecting Economic Intuition"]
E -->|"Reductive Experiment"| G["Outcome 2: Pareto Principle confirmed<br>Small portion of nodes drives market effects"]
F --> H["Key Finding: Node degree correlates with market cap<br>(ρ=0.193) & High-Influence Nodes identified"]
G --> H
H --> I["Conclusion: Simple linear models reveal inherent complexity<br>& prevent surrogate effects in crypto 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 proposes using multivariate linear VAR models to construct informative cryptocurrency networks that reflect economic intuition.
Source sentence
Here we show that multivariate linear models can create informative cryptocurrency networks that reflect economic intuition
Assumes 3
- The analysis is restricted to a VAR(1) process, assuming lag‑1 captures the causal effects. model
Source sentence
To simplify the estimation process and facilitate an analytical comparison with alternative network structures we restrict our investigation to VAR(1) processes.
- Hourly returns are computed as the logged ratio of consecutive price observations. data
Source sentence
are generated by taking the logged ratio of subsequent observations in the original price series
- Error terms in the VAR have zero mean and are often assumed Gaussian. statistical
Source sentence
vector of error terms with zero mean and some covariance matrix Σ_ε, which is often restricted to a Gaussian distribution.
Data
- Hourly cryptocurrency price series market-prices public
Source sentence
we apply the VAR framework to model the price network of 261 different cryptocurrencies, utilising hourly prices over the study period of 1/1/2020 to 1/1/2021, sourced from CoinMarketCap.com.
Costs
Not applicable
Finds 4
- Node degree is significantly correlated with market capitalisation (ρ = 0.193). statistical
Source sentence
The resulting network confirms that node degree, a measure of influence, is significantly correlated to the market capitalisation of each coin (ρ=0.193).
- Most of the network structure is concentrated in a small fraction of nodes, consistent with the Pareto principle. statistical
Source sentence
In a reductive experiment we show that most of the network structure is contained within a small portion of the network, consistent with the Pareto principle
- A subset of nodes exerts influence beyond what their market capitalisation would suggest. statistical
Source sentence
However, there remains a proportion of nodes whose influence extends beyond what their market capitalisation would imply.
- Both forms of node degree and clustering are significantly correlated with market capitalisation. statistical
Source sentence
Our multivariate model reveals significant correlation between both forms of node degree, clustering and market capitalisation
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 22 related papers). Coverage is partial, so this is neither replication nor refutation.
2 papers 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.