Paper: arXiv 2502.14431

Authors: Buddha Nath Sharma, Anish Rai, SR Luwang, Md. Nurujjaman, Sushovan Majhi

Abstract

The paper presents a comprehensive causality analysis of the US stock and commodity markets during the COVID-19 crash. The dynamics of different sectors are also compared. We use Topological Data Analysis (TDA) on multidimensional time-series to identify crashes in stock and commodity markets. The Wasserstein Distance WD shows distinct spikes signaling the crash for both stock and commodity markets. We then compare the persistence diagrams of stock and commodity markets using the WD metric. A significant spike in the \(WD\) between stock and commodity markets is observed during the crisis, suggesting significant topological differences between the markets. Similar spikes are observed between the sectors of the US market as well. Spikes obtained may be due to either a difference in the magnitude of crashes in the two markets (or sectors), or from the temporal lag between the two markets suggesting information flow. We study the Granger-causality between stock and commodity markets and also between different sectors. The results show a bidirectional Granger-causality between commodity and stock during the crash period, demonstrating the greater interdependence of financial markets during the crash. However, the overall analysis shows that the causal direction is from stock to commodity. A pairwise Granger-causal analysis between US sectors is also conducted. There is a significant increase in the interdependence between the sectors during the crash period. TDA combined with Granger-causality effectively analyzes the interdependence and sensitivity of different markets and sectors.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 3.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper employs advanced mathematics including Topological Data Analysis (TDA), persistent homology, and Vietoris–Rips complexes, which require significant theoretical understanding. However, the empirical component relies on a specific historical event (COVID-19 crash) with no mention of backtesting, live implementation, or robust out-of-sample validation, making it more of a theoretical exercise.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Analyze COVID-19 crash causality<br>in Stock & Commodity Markets"] --> B["Data: Multidimensional Time-Series<br>US Stock & Commodity Markets"]
  B --> C{"Key Methodology"}
  C --> D["TDA: Wasserstein Distance<br>Identify Crash Spikes"]
  C --> E["Granger Causality<br>Determine Interdependence"]
  D & E --> F["Computational Process:<br>Compare Persistence Diagrams<br>Test Sector/Market Causality"]
  F --> G["Key Findings:<br>1. Topological Divergence (WD Spikes)<br>2. Bidirectional Causality during Crash<br>3. Stock→Commodity info flow"]

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 detect COVID‑19‑induced crashes in US stock and commodity markets using Topological Data Analysis (TDA) and to study their interdependence with Granger‑causality analysis.

Source sentence
We use Topological Data Analysis (TDA) on multidimensional time-series to identify crashes in stock and commodity markets. We study the Granger-causality between stock and commodity markets and also between different sectors.

Assumes 5

  • The two markets have the same trading days, reducing information loss. data
    Source sentence
    Moreover, we have chosen commod-ity and stock markets since these markets have the same trading days during this period—a feature that mitigates information loss.
  • Time‑series must be stationary for Granger‑causality; non‑stationary series are differenced to achieve stationarity. statistical
    Source sentence
    If the time-series are non-stationary, differencing tech-niques may be applied to achieve stationarity37,62. III. The use of Granger-causality is appropriate when the time-series is stationary36.
  • TDA is robust to small perturbations and noise in financial data. model
    Source sentence
    Further, the method is robust to small perturbations, making TDA ideal for analyzing financial markets, which are frequently influenced by noise traders.
  • A sliding window of 30 days is appropriate for constructing point clouds for TDA. execution
    Source sentence
    In the study of stock-market dynamics, common window sizes are 20, 30, and 60 days35,49. In our analyses, we chose a window size of 30 days.
Show 1 more
  • Log‑returns of price series adequately capture market dynamics for topological analysis. model
    Source sentence
    The log-return of n stock price time-series of length l is cal-culated as Ri,j := ln Pi,j Pi,j−1 , where Pm,k denotes the closing price of the index m = {1,...,n} on day k ∈{1,...,l}. By taking the log-return of all time-series on the j-th day, i.e., (R1,j,R2,j,...,Rn,j), a point is formed in Rn.

Data

  • Three‑year US stock and commodity price data · Yahoo Finance market-prices public
    01-06-2018 to 01-06-2021 · 20 highest‑market‑cap US stocks across all major sectors and 20 commodity constituents of
    Source sentence
    DATA ANALYZED We took three years (01-06-2018 to 01-06-2021) of data from twenty US stocks with the highest market capitalization from all major sectors and twenty commodity constituents of S&P GSCI Index (ˆSPGSCI). DATA AVAILABILITY STATEMENT The data that support the findings of this study are openly available on the Yahoo Finance website at https://finance.yahoo.com67.
  • Sectoral stock data · Yahoo Finance market-prices public
    01-06-2018 to 01-06-2021 · Five highest‑market‑cap stocks from each of eleven GICS sectors
    Source sentence
    We also conducted a sectoral analysis of the US Stock market for eleven sectors based on The Global Industry Classification Standard (GICS). We se-lected the five highest market-capitalized stocks of each sector.

Costs

Not applicable The paper reports no trading or execution result.

Finds 5

  • The Wasserstein Distance shows distinct spikes during early 2020, indicating topological changes that correspond to the COVID‑19 crash in both stock and commodity markets. statistical
    Source sentence
    The Wasserstein Distance WD shows distinct spikes signaling the crash for both stock and commodity markets. A signifi-cant rise in the WDs’ is observed during early 2020 which corresponds with the onset of the COVID-19 pandemic.
  • A significant spike in the WD between stock and commodity markets is observed during the crisis, suggesting topological differences between the markets. statistical
    Source sentence
    A significant spike in the $WD$ between stock and commodity markets is observed during the crisis, suggesting significant topological differences between the markets.
  • Bidirectional Granger‑causality is detected between commodity and stock markets during the crash period, but the overall causal direction is from stock to commodity. statistical
    Source sentence
    The results show a bidirectional Granger-causality between commodity and stock during the crash period, demonstrating the greater interdependence of financial markets during the crash. However, the overall analysis shows that the causal direction is from stock to commodity.
  • Interdependence between US market sectors increases significantly during the crash period. statistical
    Source sentence
    There is a significant increase in the interdependence between the sectors during the crash period.
  • First‑difference WD1 and WD2 time‑series are stationary across all periods for both markets, enabling reliable Granger‑causality testing. statistical
    Source sentence
    Table IV shows that the p-values of both tests for the first-difference WD1 and WD2 time-series are consistently below the significance level of 0.05 for both markets in all three time periods.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 37 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 (67)

  1. Computational Topology for Data Analysis (2022)
  2. Wavelets for period analysis of unevenly sampled time series (1996)
  3. Fourier techniques for very long astrophysical time-series analysis (2002)
  4. The Fourier transform and its applications (1966)
  5. Nonlinear stochastic exclusion financial dynamics modeling and complexity behaviors (2017)
  6. Complex dynamics of financial indices (2013)
  7. The statistical mechanics of financial markets (2003)
  8. Connectedness between sectors: the case of the polish stock market before and during COVID-19 (2022)
and 59 more
  1. How COVID-19 drives connectedness among commodity and financial markets: Evidence from TVP-VAR and causality-in-quantiles techniques (2021)
  2. Sectoral connectedness: New evidence from US stock market during COVID-19 pandemics (2022)
  3. Identifying extreme events in the stock market: A topological data analysis — → 2405.16052
  4. Investigating Causal Relations by Econometric Models and Cross-spectral Methods (1969)
  5. New Introduction to Multiple Time Series Analysis (2005)
  6. Distribution of the Estimators for Autoregressive Time Series with a Unit Root (1979)
  7. Testing for a Unit Root in Time Series Regression (1988)
  8. Co-integration and Error Correction: Representation, Estimation, and Testing (1987)
  9. Linkage between international food commodity prices and the Chinese stock markets (2013)
  10. Granger causality: A review and recent advances (2022)
  11. Which lag length selection criteria should we employ? (2004)
  12. Lag-length selection and tests of Granger causality between money and income (1985)
  13. Wasserstein stability for persistence diagrams (2020) — link
  14. Causal Relationship Between Stock Market Indices and Gold Price: Evidence from India. (2013)
  15. Linear and non-linear causality between price indices and commodity prices (2014)
  16. Relationship between gold and stock markets during the global financial crisis: Evidence from nonlinear causality tests (2015)
  17. Dynamic correlations and portfolio implications across stock and commodity markets before and during the COVID-19 era: A key role of gold (2022)
  18. Does the stock market cause economic growth? Portuguese evidence of economic regime change (2013)
  19. Development of financial market and economic growth: Review of Hong Kong, China, Japan, the United States and the United Kingdom (2011)
  20. Introduction to modern time series analysis (2012)
  21. Multivariate Granger causality and generalized variance (2010)
  22. From Granger causality to long-term causality: Application to climatic data (2009)
  23. Income and total expenditure on health in OECD countries: Evidence from panel data and Hsiao’s version of Granger non-causality tests (2016)
  24. Granger causality: theory and applications (2010)
  25. Granger causality in risk and detection of extreme risk spillover between financial markets (2009)
  26. Antileakage Fourier transform for seismic data regularization in higher dimensions (2010)
  27. Critical market crashes (2003)
  28. Statistical properties of the aftershocks of stock market crashes revisited: Analysis based on the 1987 crash, financial-crisis-2008 and COVID-19 pandemic (2022)
  29. Topological data analysis of financial time series: Landscapes of crashes (2018)
  30. Detection and forecasting of extreme events in stock price triggered by fundamental, technical, and external factors (2023)
  31. Topology data analysis of critical transitions in financial networks (2017) — link
  32. Empirical study of financial crises based on topological data analysis (2020)
  33. Analysis of global stock markets’ connections with emphasis on the impact of COVID-19 (2021)
  34. Topological data analysis in investment decisions (2020)
  35. Topological tail dependence: Evidence from forecasting realized volatility (2023)
  36. Persistence of force networks in compressed granular media (2013)
  37. On time-series topological data analysis: New data and opportunities (2016)
  38. Quantifying similarity of pore-geometry in nanoporous materials (2017)
  39. Topology based data analysis identifies a subgroup of breast cancers with a unique mutational profile and excellent survival (2011)
  40. Topology data analysis using mean persistence landscapes in financial crashes (2020)
  41. Understanding changes in the topology and geometry of financial market correlations during a market crash (2021)
  42. Why topological data analysis detects financial bubbles? — → 2304.06877
  43. Topological recognition of critical transitions in time series of cryptocurrencies (2020)
  44. A sentiment-based modeling and analysis of stock price during the COVID-19: U-and Swoosh-shaped recovery (2022)
  45. Yahoo Finance (2024)
  46. Risk analysis of China’s stock markets based on topological data structures (2022)
  47. Using persistent homology as preprocessing of early warning signals for critical transition in flood (2021)
  48. Identification of short-term and long-term time scales in stock markets and effect of structural break (2020)
  49. Using topological data analysis (TDA) and persistent homology to analyze the stock markets in Singapore and Taiwan (2021)
  50. An introduction to topological data analysis: fundamental and practical aspects for data scientists (2021)
  51. An Algorithm for Persistent Homology Computation Using Homomorphic Encryption (2023) — link
  52. Stability of persistence diagrams (2005)
  53. giotto-tda:: A topological data analysis toolkit for machine learning and data exploration (2021)
  54. Financialization of commodity markets ten years later (2023)
  55. A Shock Like No Other: The Impact of Covid-19 on Commodity Markets (2020)
  56. COVID-19 and the march 2020 stock market crash. Evidence from S&P1500 (2021)
  57. Impact of the COVID-19 outbreak on the US equity sectors: Evidence from quantile return spillovers (2021)
  58. Least-squares wavelet analysis and its applications in geodesy and geophysics (2018)
  59. COVID-19 pandemic, economic indicators and sectoral returns: evidence from US and China (2022)

2 of 67 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.