Paper: arXiv 2502.07625
Authors: S. R. Luwang, A. Rai, Md. Nurujjaman, F. Petroni
Abstract
An empirical stochastic analysis of high-frequency, tick-by-tick order data of NASDAQ100 listed stocks is conducted using a first-order discrete-time Markov chain model to explore intraday order transition dynamics. This analysis focuses on three market cap categories: High, Medium, and Low. Time-homogeneous transition probability matrices are estimated and compared across time-zones and market cap categories, and we found that limit orders exhibit higher degree of inertia (DoI), i.e., the probability of placing consecutive limit order is higher, during the opening hour. However, in the subsequent hour, the DoI of limit order decreases, while that of market order increases. Limit order adjustments via additions and deletions of limit orders increases significantly after the opening hour. All the order transitions then stabilize during mid-hours. As the closing hour approaches, consecutive order executions surge, with decreased placement of buy and sell limit orders following sell and buy executions, respectively. In terms of the differences in order transitions between stocks of different market cap, DoI of orders is stronger in high and medium market cap stocks. On the other hand, lower market cap stocks show a higher probability of limit order modifications and greater likelihood of submitting sell/buy limit orders after buy/sell executions. Further, order transitions are clustered across all stocks, except during opening and closing hours. The findings of this study may be useful in understanding intraday order placement dynamics across stocks of varying market cap, thus aiding market participants in making informed order placements at different times of trading hour.
Complexity vs Empirical Score
- Math Complexity: 5.0/10
- Empirical Rigor: 8.0/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: The paper employs a Markov chain model, which involves matrix algebra and probability theory, placing it in the medium-high math complexity range. Empirically, it uses high-frequency tick-by-tick order data from NASDAQ100 stocks, conducts statistical analysis with transition matrices and Jensen-Shannon divergence, and presents findings on order dynamics, indicating strong data/implementation-heavy rigor.
Research Flowchart
flowchart TD
Start(["Research Goal: Analyze Intraday Order Transition Dynamics"]) --> Data["Data Source: High-Frequency Tick-by-Tick Data<br>NASDAQ100 Stocks"]
Data --> Category{"Categorize Stocks by<br>Market Cap"}
Category --> H["High Market Cap"]
Category --> M["Medium Market Cap"]
Category --> L["Low Market Cap"]
H --> Method
M --> Method
L --> Method
Method["Methodology: First-Order<br>Discrete-Time Markov Chain"] --> Compute["Compute Time-Homogeneous<br>Transition Probability Matrices"]
Compute --> Analysis["Analysis: Compare DoI & Transitions<br>across Time Zones & Cap Categories"]
Analysis --> Findings{"Key Findings"}
Findings --> F1["Inertia (DoI) peaks at opening<br>(Limit Orders dominant)"]
Findings --> F2["Subsequent hour: DoI shifts<br>(Limit decreases, Market increases)"]
Findings --> F3["Mid-hours: Transitions stabilize"]
Findings --> F4["Closing hour: Execution surge<br>(Limit placement drops)"]
Findings --> F5["Cap Differences: High/Med have<br>stronger DoI; Low have more<br>modifications/executions"]
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 model to analyze intraday order transition dynamics across high, medium and low market‑cap NASDAQ100 stocks.
Source sentence
In the present study, we use a first-order time-homogeneous discrete time Markov chain (DTMC) model to examine the intraday order transition dynamics.
Assumes 5
- Order transitions follow a first‑order Markov property, i.e., dependence only on the immediately preceding event. model
Source sentence
In this study, we focus only on the dependence between consecutive events, thus enabling the application of a first-order Markov property to high-frequency tick-by-tick order sequence data.
- Transition probabilities are time‑homogeneous within each defined time‑zone. model
Source sentence
An empirical stochastic analysis of high-frequency, tick-by-tick order data of NASDAQ100 listed stocks is conducted using a first-order discrete-time Markov chain model to explore intraday order transition dynamics.
- The Markov chains are finite, ergodic (aperiodic and irreducible), allowing a unique stationary distribution. statistical
Source sentence
For a finite and ergodic (i.e., aperiodic and irreducible) Markov chain, the transition probabilities stabilize to this distribution as the number of steps increases indefinitely51.
- Order types listed in Table II constitute the state space of the Markov chain. data
Source sentence
Considering these order types as states of the Markov chain, we will analyze the transition probabilities between them in this study.
Show 1 more
- High‑frequency tick‑by‑tick order data from Algoseek accurately captures all order events from 04:00 to 20:00 EST. data
Source sentence
For this study, Algoseek40 provided the high-frequency tick-by-tick order transaction data. The provided full data format contains information of all the types of orders placed starting from 04:00:00 EST to 20:00:00 EST for all the listed stocks in NASDAQ100.
Data
- Algoseek high‑frequency tick‑by‑tick order transaction data order-book commercial
Source sentence
For this study, Algoseek40 provided the high-frequency tick-by-tick order transaction data. The provided full data format contains information of all the types of orders placed starting from 04:00:00 EST to 20:00:00 EST for all the listed stocks in NASDAQ100.
Costs
Not applicable
Finds 5
- Limit orders exhibit a higher degree of inertia (DoI) during the opening hour, meaning consecutive limit order submissions are more likely. statistical
Source sentence
Time-homogeneous transition probability matrices are estimated and compared across time-zones and market cap categories, and we found that limit orders exhibit higher degree of inertia (DoI), i.e., the probability of placing consecutive limit order is higher, during the opening hour.
- In the second hour, DoI of limit orders decreases while DoI of market orders increases. statistical
Source sentence
However, in the subsequent hour, the DoI of limit order decreases, while that of market order increases.
- Overall DoI is larger for high‑ and medium‑cap stocks than for low‑cap stocks, indicating more consecutive identical order submissions in larger caps. statistical
Source sentence
In terms of the differences in order transitions between stocks of different market cap, DoI of orders is stronger in high and medium market cap stocks. 2. Differences between HMC, MMC and LMC in order transition dynamics [Refer to Fig. 1]: The overall DoI of the orders is larger for HMC and MMC as compared to LMC stocks, which indicates that consecutive submissions of identical orders are more prevalent in HMC and
- Low‑cap stocks show a higher probability of limit order modifications (additions/deletions) and a greater likelihood of submitting opposite‑side limit orders after executions. statistical
Source sentence
On the other hand, lower market cap stocks show a higher probability of limit order modifications and greater likelihood of submitting sell/buy limit orders after buy/sell executions. This indicates that the probability of immediate addition after deletion of a limit order is more prominent as compared to the probability of immediate deletion after addition for lower market cap stocks.
- Order transition dynamics cluster across all stocks except during opening and closing hours, where distinct patterns emerge. statistical
Source sentence
Further, order transitions are clustered across all stocks, except during opening and closing hours. This indicates that the order transition dynamics from 10:30 to 2:00 are similar, irrespective of the market cap.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 21 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 (69)
- The probability of limit-order execution (2000)
- A semi-Markov model with memory for price changes (2011)
- A new methodology for studying intraday dynamics of Nikkei index futures using Markov chains (1999)
- Price dynamics in limit order markets (1998)
- Stochastic Conditional Duration Model with Intraday Seasonality and Limit Order Book Information (2022)
- Order submission: The choice between limit and market orders (2003)
- Resiliency of the limit order book (2015)
- Analysis of symbolic sequences using the Jensen-Shannon divergence (2002)
and 61 more
- Detecting dynamical changes in time series by using the Jensen Shannon divergence (2017)
- Urban flood risk assessment based on DBSCAN and K-means clustering algorithm (2023)
- Spatio-temporal modeling of seismic provinces of Iran using DBSCAN algorithm (2017)
- Integrating independent component analysis and principal component analysis with neural network to predict Chinese stock market (2011)
- Enhanced Fault Detection in Energy Systems Using Individual Contextual Forgetting Factors in Recursive Principal Component Analysis (2024)
- Measuring the Information Content of Stock Trades (1991)
- The diversity of high-frequency traders (2013)
- Detection and forecasting of extreme events in stock price triggered by fundamental, technical, and external factors (2023)
- A simple introduction to Markov Chain Monte–Carlo sampling (2018)
- Thermodynamics and Statistical Mechanics (2020)
- Scikit-learn: Machine learning in Python (2011)
- Algoseek.com (2023) — link
- PyDTMC (2024) — link
- Data-driven stochastic modeling Markov Chain Matlab Codes (2024) — link
- Discrete-time Markov chain modelling of the ontario air quality health index (2021)
- Markov Chain Modelling in Finance: Stock Valuation and Price Discovery (2019)
- A Markov Chain Model of Air Quality Index: Modelling and Simulation (2020)
- The markovchain package: a package for easily handling Discrete Markov Chains in R (2016)
- First and second order Markov chain models for synthetic generation of wind speed time series (2005)
- Markov Chains based route travel time estimation considering link spatio-temporal correlation (2020)
- Markov chain model for the stochastic behaviors of wind-direction data (2015)
- Statistical inference about Markov chains (1957)
- Stochastic processes (1994)
- Analyzing Sequence Data with Markov Chain Models in Scientific Experiments (2021)
- Applicability of G-test in analyzing categorical variables (2019)
- USP: an independence test that improves on Pearson’s chi-squared and the G-test (2021)
- Accelerating causal inference and feature selection methods through G-test computation reuse (2021)
- A Dynamic Model of Order Execution and the Intraday Cost of Limit Orders (2004)
- On the Jensen–Shannon symmetrization of distances relying on abstract means (2019)
- Liquidity supply and adverse selection in a pure limit order book market (2008)
- Measuring the information content of stock trades (1991)
- Order flow composition and trading costs in a dynamic limit order market (1999)
- Liquidity, information, and infrequently traded stocks (1996)
- Intraday trading patterns in an intelligent autonomous agent-based stock market (2011)
- High-frequency trading and stock liquidity: An intraday analysis (2020)
- Order book dynamics in liquid markets: limit theorems and diffusion approximations (2012) — link
- Transactions costs and investment style: an inter-exchange analysis of institutional equity trades (1997)
- High-frequency stock market order transitions during the US–China trade war 2018: A discrete-time Markov chain analysis — → 2405.05634
- The Holt-winters forecasting procedure (1978)
- Statistical methods for data with long-range dependence (1992)
- An investigation of transactions data for NYSE stocks (1985)
- Market vs. limit orders: the SuperDOT evidence on order submission strategy (1996)
- Optimal dynamic order submission strategies in some stylized trading problems (1998)
- Cancellation and uncertainty aversion on limit order books (2004)
- Behaviour of intra-daily stock return on an Asian emerging market-Hong Kong (1991)
- Spreads, depths, and the impact of earnings information: An intraday analysis (1993)
- A theory of intraday patterns: Volume and price variability (1988)
- Periodic market closure and trading volume: A model of intraday bids and asks (1992)
- Limit orders and the bid–ask spread (1999)
- Market structure and the intraday pattern of bid-ask spreads for NASDAQ securities (1995)
- An analysis of depth behavior in an electronic, order-driven environment (1999)
- Periodic market closure and trading volume: A model of intraday bids and asks (1992)
- Order dynamics: Recent evidence from the NYSE (2007)
- An empirical analysis of the limit order book and the order flow in the Paris Bourse (1995)
- Information flows and open outcry: Evidence of imitation trading (1998)
- Order submission strategy and the curious case of marketable limit orders (2002)
- So what orders do informed traders use? (2006)
- The deadline effect in bargaining: Some experimental evidence (1988)
- Empirical analysis of limit order markets (2004)
- Intraday liquidity costs and order execution quality in NASDAQ stocks (2007)
- Intraday time and order execution quality dimensions (2009)
1 of 69 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.