Paper: arXiv 2308.08135
Authors: Xianfeng Jiao, Zizhong Li, Chang Xu, Yang Liu, Weiqing Liu, Jiang Bian
Abstract
High-frequency quantitative investment is a crucial aspect of stock investment. Notably, order flow data plays a critical role as it provides the most detailed level of information among high-frequency trading data, including comprehensive data from the order book and transaction records at the tick level. The order flow data is extremely valuable for market analysis as it equips traders with essential insights for making informed decisions. However, extracting and effectively utilizing order flow data present challenges due to the large volume of data involved and the limitations of traditional factor mining techniques, which are primarily designed for coarser-level stock data. To address these challenges, we propose a novel framework that aims to effectively extract essential factors from order flow data for diverse downstream tasks across different granularities and scenarios. Our method consists of a Context Encoder and an Factor Extractor. The Context Encoder learns an embedding for the current order flow data segment’s context by considering both the expected and actual market state. In addition, the Factor Extractor uses unsupervised learning methods to select such important signals that are most distinct from the majority within the given context. The extracted factors are then utilized for downstream tasks. In empirical studies, our proposed framework efficiently handles an entire year of stock order flow data across diverse scenarios, offering a broader range of applications compared to existing tick-level approaches that are limited to only a few days of stock data. We demonstrate that our method extracts superior factors from order flow data, enabling significant improvement for stock trend prediction and order execution tasks at the second and minute level.
Complexity vs Empirical Score
- Math Complexity: 6.5/10
- Empirical Rigor: 8.0/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: The paper employs advanced unsupervised learning, attention mechanisms, and information theory, indicating high mathematical complexity, while it demonstrates empirical rigor by using a full year of tick-level data across multiple scenarios and validating on real downstream tasks like trend prediction and order execution.
Research Flowchart
flowchart TD A["Research Goal<br>Extract & Utilize<br>Order Flow Factors"] --> B["Data Input<br>High-Frequency<br>Order Flow Data"] B --> C["Methodology: Context Encoder<br>Learn Embedding of<br>Expected vs Actual Market State"] B --> D["Methodology: Factor Extractor<br>Unsupervised Learning to<br>Select Distinct Signals"] C --> E["Computational Process<br>Integration of Context<br>& Extracted Factors"] D --> E E --> F["Outcome: Superior Factors<br>For Downstream Tasks"] F --> G["Findings<br>Improved Stock Trend<br>Prediction & Order Execution"]
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 novel framework with a Context Encoder and a Factor Extractor to extract essential factors from order flow data for downstream tasks.
Source sentence
we propose a novel framework that aims to effectively extract essential factors from order flow data for diverse downstream tasks across different granularities and scenarios.
Assumes 1
- The order book is updated instantaneously when a new order is received. model
Source sentence
To simplify notation and analysis, we assume that the order book is updated instantaneously when a new order 𝑥𝑡+1 is received.
Data
- order flow dataset order-book unknown
Source sentence
Our dataset consists of two parts: candlestick (at the daily frequency) as baseline daily factor extractor dataset, and high-frequency (10‑ms) order flow data as our intra‑day micro‑factor extractor dataset. Our dataset is collected from the 10 most active stocks of CSI300.
Yahoo Finance
Costs
Not modeled
Finds 2
- The method achieves superior performance compared to other microstructure‑based factor mining methods in both daily return prediction and order execution downstream tasks. performance
Source sentence
it achieves superior performance compared to other microstructure-based factor mining methods in both daily return prediction and order execution downstream tasks.
- The proposed machine‑learning‑based method improves order execution performance by 38.64% in PA and 3% in GLR over the best rule‑based baseline. performance
Source sentence
our proposed machine learning‑based method exhibits an improvement of 38.64% in PA, and an improvement 3% in GLR, demonstrating our method’s effectiveness in optimizing the order execution task.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 7 related papers). Coverage is partial, so this is neither replication nor refutation.
2 papers in the archive : From Deep Learning to LLMs: A survey of AI in Quantitative …
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.