Paper: arXiv 2405.05634

Authors: Salam Rabindrajit Luwang, Anish Rai, Md. Nurujjaman, Om Prakash, Chittaranjan Hens

Abstract

Statistical analysis of high-frequency stock market order transaction data is conducted to understand order transition dynamics. We employ a first-order time-homogeneous discrete-time Markov chain model to the sequence of orders of stocks belonging to six different sectors during the USA-China trade war of 2018. The Markov property of the order sequence is validated by the Chi-square test. We estimate the transition probability matrix of the sequence using maximum likelihood estimation. From the heat-map of these matrices, we found the presence of active participation by different types of traders during high volatility days. On such days, these traders place limit orders primarily with the intention of deleting the majority of them to influence the market. These findings are supported by high stationary distribution and low mean recurrence values of add and delete orders. Further, we found similar spectral gap and entropy rate values, which indicates that similar trading strategies are employed on both high and low volatility days during the trade war. Among all the sectors considered in this study, we observe that there is a recurring pattern of full execution orders in Finance & Banking sector. This shows that the banking stocks are resilient during the trade war. Hence, this study may be useful in understanding stock market order dynamics and devise trading strategies accordingly on high and low volatility days during extreme macroeconomic events.

Complexity vs Empirical Score

  • Math Complexity: 6.0/10
  • Empirical Rigor: 3.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper employs discrete-time Markov chains, spectral analysis, and statistical hypothesis testing (Chi-square), indicating moderate mathematical complexity. However, it lacks implementation details, backtests, or specific datasets, and the empirical findings are presented as theoretical observations rather than actionable trading results.

Research Flowchart

  flowchart TD
  A["Research Goal: Analyze HFT order transitions<br>during 2018 US-China Trade War"] --> B["Data: High-frequency order book data<br>from 6 sectors"]
  B --> C["Methodology: Discrete-Time Markov Chain<br>+ Chi-square test + MLE"]
  C --> D{"Computational Processes"}
  D --> E["Estimate Transition Matrices"]
  D --> F["Calculate Stationary Distributions"]
  D --> G["Compute Spectral Gap & Entropy"]
  E --> H["Findings"]
  F --> H
  G --> H
  H --> I["Active participation by traders on high volatility days"]
  H --> J["Similar strategies used on both high/low volatility days"]
  H --> K["Finance & Banking sector shows resilience<br>(recurring full execution orders)"]

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 first-order time‑homogeneous discrete‑time Markov chain (DTMC) model to analyze high‑frequency stock market order transition dynamics during the 2018 US‑China trade war.

Source sentence
We employ a first-order time-homogeneous discrete-time Markov chain model to the sequence of orders of stocks belonging to six different sectors during the USA-China trade war of 2018.

Assumes 6

  • The order sequence follows the Markov property, i.e., the next order depends only on the current order. statistical
    Source sentence
    The Markov property of the order sequence is validated by the Chi-square test.
  • The Markov chain is time‑homogeneous (transition probabilities do not change over time). model
    Source sentence
    We employ a first-order time-homogeneous discrete-time Markov chain model to the sequence of orders of stocks belonging to six different sectors during the USA-China trade war of 2018.
  • The chain is stationary, allowing the use of stationary distribution and entropy‑rate formulas. model
    Source sentence
    The entropy rate of the process is as follows: H(X) = lim n→∞H(Xn | Xn−1, Xn−2, ...., X0) = lim n→∞H(Xn | Xn−1) = lim n→∞ X si∈S Pr(Xn−1 = si)H(Xn | Xn−1 = si) = −lim n→∞ X si∈S Pr(Xn−1 = si) X sj∈S Pr(Xn = sj | Xn−1 = si) log2 Pr(Xn = sj | Xn−1 = si) (16) The stationary assumption and an assumption of a time-homogeneous Markov process imply Pr(Xn = sj | Xn−1 = si) = p(si, sj) = pij.
  • The DTMC for each stock is ergodic and irreducible. model
    Source sentence
    The Markov chain for each stock representing the corresponding sectors is found to be ergodic and irreducible.
Show 2 more
  • High‑frequency order data from Algoseek accurately represent the true order flow for the selected 18 stocks. data
    Source sentence
    DATA DESCRIPTION For this study, Algoseek60 provided high-frequency stock market order transaction data. From this 10s of millions long data which is for all the listed stocks, we extract the data for 18 stocks, belonging to 6 sectors as shown in VII, for this study.
  • Selected stocks are the three largest‑market‑cap stocks in each of six sectors. data
    Source sentence
    A. Selection of High and Low volatility days In this study, we have selected three active stocks listed in NASDAQ from each of the six sectors. The selection of the stocks was based on the largest market capitalization, as presented in Table VII of the Appendix.

Data

  • Algoseek high‑frequency order transaction data · Algoseek order-book commercial
    2018-11-01 to 2018-12-31 · 18 NASDAQ stocks from six sectors
    Source sentence
    DATA DESCRIPTION For this study, Algoseek60 provided high-frequency stock market order transaction data. We have analysed the data from November 1st, 2018 to December 31st, 2018.

Costs

Not applicable The paper reports no trading or execution result.

Finds 5

  • The probability of placing limit orders immediately after full execution (FB→AA and FA→AB) is higher on high‑volatility days than on low‑volatility days. statistical
    Source sentence
    14 Our findings reveal that the probability of placing limit orders immediately following the complete execution of an order, i.e., FB →AA and FA →AB, are high on high volatility days compared to low volatility days.
  • Transition probabilities from add orders to delete orders (AB→DB and AA→DA) are very high, indicating traders place limit orders primarily to delete them on high‑volatility days. statistical
    Source sentence
    We also found that the transition probability from an add orders to a delete orders, i.e., AB →DB and AA →DA, are very high, which suggests that traders place limit orders primarily with the intention of deleting the majority of them to influence the market on high volatility days.
  • A recurring pattern of full execution of bid and ask orders (FB and FA) is observed in the order sequence of banking stocks. statistical
    Source sentence
    The result also shows a recurring pattern of the full execution of bid and ask orders, i.e., FB and FA, in the order sequence of banking stocks.
  • Finance & Banking sector shows a small spectral gap and entropy rate with a large relaxation rate, confirming the recurring execution pattern. statistical
    Source sentence
    Small spectral gap and entropy rate and a large relaxation rate of Finance & Banking sector confirms recurring the pattern.
  • Full‑execution orders FB and FA have a high mean recurrence time, indicating they occur far apart in the sequence. statistical
    Source sentence
    A high mean recurrence time (MRT) for FB and FA also indicates that full order execution occurs very far apart.

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.

Cited by 6 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.

Cited References (79)

  1. Multiple duration analyses of dynamic limit order placement strategies and aggressiveness in a low-latency market environment (2020)
  2. Information, liquidity, and dynamic limit order markets (2022)
  3. Identification of short-term and long-term time scales in stock markets and effect of structural break (2020)
  4. Time scales and characteristics of stock markets in different investment horizons (2020)
  5. Characteristics of 2020 stock market crash: The COVID-19 induced extreme event (2021)
  6. Limit order revisions (2010)
  7. The flash crash: High-frequency trading in an electronic market (2017)
  8. Low-latency trading (2013)
and 71 more
  1. Automation, speed, and stock market quality: The NYSE’s hybrid (2011)
  2. The diversity of high-frequency traders (2013)
  3. Day trading and stock price volatility (2008)
  4. Trading and exchanges: Market microstructure for practitioners (2003)
  5. Modeling and analysis of the effect of COVID-19 on the stock price: V and L-shape recovery (2021)
  6. Statistical properties of the aftershocks of stock market crashes revisited: Analysis based on the 1987 crash, financial-crisis-2008 and COVID-19 pandemic (2022)
  7. A sentiment-based modeling and analysis of stock price during the COVID-19: U-and Swoosh-shaped recovery (2022)
  8. Detection and forecasting of extreme events in stock price triggered by fundamental, technical, and external factors (2023)
  9. COVID-19 and stock market volatility: An industry level analysis (2020)
  10. The effect of COVID–19 pandemic on global stock market volatility: can economic strength help to manage the uncertainty? (2021)
  11. Policy uncertainty and sectoral stock market volatility in China (2021)
  12. Sentiment and stock market connectedness: Evidence from the US–China trade war (2022)
  13. Market Volatility Revs Up Amid Trade War Tensions (2018)
  14. Does the US-China trade war affect co-movements between US and Chinese stock markets? (2021)
  15. The impact of the US–China trade war on the European Union (2020)
  16. Costs and consequences of a trade war: a structural analysis (2018)
  17. What impacts foreign capital flows to China’s stock markets? Evidence from financial risk spillover networks (2023)
  18. Measuring Dependence in China-United States Trade War: A Dynamic Copula Approach for BRICV and US Stock Markets (2021)
  19. Effect of the US–China Trade War on Stock Markets: A Financial Contagion Perspective (2021) — link
  20. Statistical modeling of high-frequency financial data (2011)
  21. A cube model for web access sessions and cluster analysis (2001)
  22. Higher-order Markov chain models for categorical data sequences (2004)
  23. moveHMM: an R package for the statistical modelling of animal movement data using hidden Markov models (2016)
  24. Analysis of movements and behavior of caribou (Rangifer tarandus) using hidden Markov models (2004)
  25. Hidden Markov models for circular and linear-circular time series (2006)
  26. A multivariate Markov chain model for categorical data sequences and its applications in demand predictions (2002)
  27. A simple introduction to Markov Chain Monte–Carlo sampling (2018)
  28. Thermodynamics and Statistical Mechanics (2020)
  29. Continuous time Markov chain models for chemical reaction networks (2011)
  30. FOG: Fragment Optimized Growth algorithm for the de novo generation of molecules occupying druglike chemical space (2009)
  31. Towards a mathematical theory of cortical micro-circuits (2009)
  32. Markov analysis in human resource administration: Applications and limitations (1977)
  33. Networks in climate (2019)
  34. Control techniques for complex networks (2008)
  35. Retail banking loan portfolio equilibrium mix: A Markov chain model analysis (2005)
  36. On a multivariate Markov chain model for credit risk measurement (2005)
  37. Estimating Markov transition matrices using proportions data: an application to credit risk (2005)
  38. A semi-Markov model for price returns (2012)
  39. A semi-Markov model with memory for price changes (2011)
  40. Modeling the conditional distribution of interest rates as a regime-switching process (1996)
  41. Using time-varying transition probabilities in Markov switching processes to adjust US fiscal policy for asset prices (2013)
  42. Transitions in the stock markets of the US, UK and Germany (2017)
  43. Modelling non-linear economic relationships (1993)
  44. The euro’s impacts on the smooth transition dynamics of stock market volatilities (2012)
  45. Modeling tools to evaluate the performance of wireless multi-hop networks (2015)
  46. Performance modeling of communication networks with Markov chains (2010)
  47. An higher-order Markov chain model for prediction of categorical data sequences (2003)
  48. Video-based face recognition using adaptive hidden markov models (2003)
  49. Holdings of financial assets: A Markov chain analysis (1982)
  50. First order Markov chain model and rainfall sequences in several stations of Spain (2001)
  51. Fitting models to daily rainfall data (1982)
  52. Algoseek.com (2023) — link
  53. Discrete-time Markov chain modelling of the ontario air quality health index (2021)
  54. Markov Chain Modelling in Finance: Stock Valuation and Price Discovery (2019)
  55. A Markov Chain Model of Air Quality Index: Modelling and Simulation (2020)
  56. The markovchain package: a package for easily handling Discrete Markov Chains in R (2016)
  57. First and second order Markov chain models for synthetic generation of wind speed time series (2005)
  58. Markov Chains based route travel time estimation considering link spatio-temporal correlation (2020)
  59. Markov chain models for vegetation dynamics (2000)
  60. Markov chain model for the stochastic behaviors of wind-direction data (2015)
  61. Application of Cellular automata and Markov-chain model in geospatial environmental modeling-A review (2017)
  62. Statistical inference about Markov chains (1957)
  63. Markov chains (2006)
  64. What is fundamental for Markov chains: First passage times, fundamental matrices, and group generalized inverses (1995)
  65. Stochastic processes (1994)
  66. Non-negative matrices and Markov chains (2006)
  67. Elements of information theory (1999)
  68. Estimating the entropy rate of finite Markov chains with application to behavior studies (2019)
  69. On the estimation of the entropy rate of finite Markov chains (2005)
  70. Analyzing Sequence Data with Markov Chain Models in Scientific Experiments (2021)
  71. High-frequency trading: Order-based innovation or manipulation? (2020)

0 of 79 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.