Understanding the temporal stochastic dynamics of order submissions is essential to market microstructure research and the design of robust trading strategies. This study employs a Markov chain framework to empirically analyze high-frequency tick-by-tick order submission data from NASDAQ100 stocks, exploring divergent intraday order transition patterns across High (HMC), Medium (MMC), and Low (LMC) market capitalization groups. Elementwise comparisons of transition probability matrices (TPMs) reveal: (1) limit orders exhibit peak Degree of Inertia (DoI) at market open, shifting to market orders in the subsequent hour alongside increased limit order modifications; (2) order flows stabilize mid-day but aggressive executions dominate the closing hour with reduced limit order placements; and (3) HMC stocks show stronger DoI and buying dominance, while LMC stocks display higher limit order modifications and greater post-execution limit order submissions (sell/buy after buy/sell). Dimensionality reduction via Principal component analysis and t-distributed stochastic neighbor embedding, followed by Hierarchical and density-based clustering of TPMs of different time intervals confirms three universal intraday phases—opening, mid-day, and closing, irrespective of the market capitalization. Furthermore, stationary distributions comparison through Jensen–Shannon divergence further demonstrate mid-day similarity across market capitalization groups, but divergent behaviors at critical hours: MMC stocks show unique opening hour dynamics, while HMC/LMC stocks exhibit distinct signatures in closing hour. In summary, this paper contributes a data-driven framework integrating Markov chains, dimensionality-reduced clustering, and stationary analysis to decode stochastic order submission dynamics. It provides actionable insights on order submissions across market capitalizations and trading hours, enabling robust trading strategies that mitigate intraday trading risks. © The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2025.
Intraday order transitions across market capitalizations through Markov analysis
Petroni Filippo
2025-01-01
Abstract
Understanding the temporal stochastic dynamics of order submissions is essential to market microstructure research and the design of robust trading strategies. This study employs a Markov chain framework to empirically analyze high-frequency tick-by-tick order submission data from NASDAQ100 stocks, exploring divergent intraday order transition patterns across High (HMC), Medium (MMC), and Low (LMC) market capitalization groups. Elementwise comparisons of transition probability matrices (TPMs) reveal: (1) limit orders exhibit peak Degree of Inertia (DoI) at market open, shifting to market orders in the subsequent hour alongside increased limit order modifications; (2) order flows stabilize mid-day but aggressive executions dominate the closing hour with reduced limit order placements; and (3) HMC stocks show stronger DoI and buying dominance, while LMC stocks display higher limit order modifications and greater post-execution limit order submissions (sell/buy after buy/sell). Dimensionality reduction via Principal component analysis and t-distributed stochastic neighbor embedding, followed by Hierarchical and density-based clustering of TPMs of different time intervals confirms three universal intraday phases—opening, mid-day, and closing, irrespective of the market capitalization. Furthermore, stationary distributions comparison through Jensen–Shannon divergence further demonstrate mid-day similarity across market capitalization groups, but divergent behaviors at critical hours: MMC stocks show unique opening hour dynamics, while HMC/LMC stocks exhibit distinct signatures in closing hour. In summary, this paper contributes a data-driven framework integrating Markov chains, dimensionality-reduced clustering, and stationary analysis to decode stochastic order submission dynamics. It provides actionable insights on order submissions across market capitalizations and trading hours, enabling robust trading strategies that mitigate intraday trading risks. © The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2025.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
