Paper: arXiv 2310.18903
Authors: Yan-Hong Yang, Ying-Lin Liu, Ying-Hui Shao
Abstract
Drawing inspiration from the significant impact of the ongoing Russia-Ukraine conflict and the recent COVID-19 pandemic on global financial markets, this study conducts a thorough analysis of three key crude oil futures markets: WTI, Brent, and Shanghai (SC). Employing the visibility graph (VG) methodology, we examine both static and dynamic characteristics using daily and high-frequency data. We identified a clear power-law decay in most VG degree distributions and highlighted the pronounced clustering tendencies within crude oil futures VGs. Our results also confirm an inverse correlation between clustering coefficient and node degree and further reveal that all VGs not only adhere to the small-world property but also exhibit intricate assortative mixing. Through the time-varying characteristics of VGs, we found that WTI and Brent demonstrate aligned behavior, while the SC market, with its unique trading mechanics, deviates. The 5-minute VGs’ assortativity coefficient provides a deeper understanding of these markets’ reactions to the pandemic and geopolitical events. Furthermore, the differential responses during the COVID-19 and Russia-Ukraine conflict underline the unique sensitivities of each market to global disruptions. Overall, this research offers profound insights into the structure, dynamics, and adaptability of these essential commodities markets in the face of worldwide challenges.
Complexity vs Empirical Score
- Math Complexity: 7.5/10
- Empirical Rigor: 8.0/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: The paper employs advanced network theory concepts like visibility graphs, power-law distributions, and small-world properties, requiring dense mathematical formalism. It also demonstrates high empirical rigor by analyzing real high-frequency and daily data from multiple crude oil futures markets (WTI, Brent, Shanghai) across specific crisis periods (COVID-19, Russia-Ukraine conflict) and reporting detailed statistical metrics (degree distributions, clustering coefficients, assortativity).
Research Flowchart
flowchart TD
A["Research Goal<br>Analyze crude oil futures market dynamics<br>under COVID-19 & Russia-Ukraine conflict"] --> B["Methodology<br>Visibility Graph (VG) Analysis"]
B --> C{"Data Inputs<br>Daily & 5-minute futures prices<br>WTI, Brent, Shanghai SC"}
C --> D["Computational Process<br>Construct Static & Dynamic VGs"]
D --> E{"Analyze Network Properties<br>Degree Distribution<br>Clustering Coefficient<br>Assortativity"}
E --> F["Key Findings<br>Power-law decay & Small-World property<br>WTI & Brent aligned; SC deviates<br>Unique sensitivities to global shocks"]
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 to explore the effects of crisis events on crude oil futures markets through the lens of network topology evolution using the visibility graph method.
Source sentence
In this study, our primary contribution is to explore the effects of crisis events on crude oil futures markets through the lens of network topology evolution.
Assumes 4
- Front‑month futures contracts are representative of the continuous price series for each crude oil market. data
Source sentence
We use the front-month futures contracts to form continuous closing price time series for all three assets.
- The visibility graph method can capture the static and dynamic characteristics of financial time series. model
Source sentence
It’s worth noting that the visibility graph method we employed was originally introduced by Lacasa et al. 2 (2008), serving as an innovative approach to time series analysis within the realm of complex networks.
- High‑frequency intervals (5‑, 15‑, 30‑minute) are sufficient to minimize noise in the price series. data
Source sentence
Consistent with many studies, we choose the 5-minute, 15-minute, and 30-minute frequencies to minimize noise (Shao et al., 2023).
- All three crude oil futures series are non‑stationary, justifying the use of visibility graphs on raw price levels. statistical
Source sentence
Furthermore, the Augmented Dickey-Fuller (ADF) tests indicate that all series are nonstationary.
Data
- WTI, Brent, and Shanghai (SC) crude oil futures price series market-prices commercial
Source sentence
Data description and summary statistics Throughout this research, we select WTI, Brent, and SC as the three benchmark indices for global crude oil futures. Our data was sourced from the Wind database.
Costs
Not applicable
Finds 5
- Most visibility graphs exhibit a clear power‑law decay in degree distributions, with scaling parameter α between 2 and 3.5. statistical
Source sentence
We identified a clear power-law decay in most VG degree distributions and highlighted the pronounced clustering tendencies within crude oil futures VGs. Most VGs demonstrated a clear power-law decay in their degree distributions, with the scaling parameter α varying between 2 and 3.5.
- An inverse relationship is observed between clustering coefficient and node degree across the crude oil futures VGs. statistical
Source sentence
Our results also confirm an inverse correlation between clustering coefficient and node degree and further reveal that all VGs not only adhere to the small-world property but also exhibit intricate assortative mixing. We recognized the pronounced clustering tendencies in crude oil futures VGs and verified the inverse relationship between the clustering coefficient and node degree.
- All VGs satisfy the small‑world property and display complex assortative mixing patterns. statistical
Source sentence
Our results also confirm an inverse correlation between clustering coefficient and node degree and further reveal that all VGs not only adhere to the small-world property but also exhibit intricate assortative mixing. Furthermore, the assortativity coefficient r and the average nearest neighbor degree ⟨knn|k⟩together indicated that VGs, irrespective of their frequency, display complex assortative mixing patterns.
- WTI and Brent markets show aligned dynamic behavior, while the Shanghai market deviates due to its unique trading mechanics. other
Source sentence
Through the time-varying characteristics of VGs, we found that WTI and Brent demonstrate aligned behavior, while the SC market, with its unique trading mechanics, deviates. The closely mirroring patterns between WTI and Brent suggest their behavioral alignment, while SC’s unique dynamics can be attributed to its distinct trading procedures.
- The 5‑minute assortativity coefficient effectively reflects the short‑term market responses to the COVID‑19 pandemic and the Russia‑Ukraine conflict. statistical
Source sentence
The 5-minute VGs' assortativity coefficient provides a deeper understanding of these markets' reactions to the pandemic and geopolitical events. Specifically, the 5-minute VGs’ assortativity coefficient r stands out in reflecting the major impacts of the COVID-19 pandemic and the Russia-Ukraine conflict on the crude oil futures markets.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 23 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.