Paper: arXiv 2404.12001
Authors: Peng Yifeng
Abstract
Market fluctuations caused by overtrading are important components of systemic market risk. This study examines the effect of investor sentiment on intraday overtrading activities in the Chinese A-share market. Employing high-frequency sentiment indices inferred from social media posts on the Eastmoney forum Guba, the research focuses on constituents of the CSI 300 and CSI 500 indices over a period from 01/01/2018, to 12/30/2022. The empirical analysis indicates that investor sentiment exerts a significantly positive impact on intraday overtrading, with the influence being more pronounced among institutional investors relative to individual traders. Moreover, sentiment-driven overtrading is found to be more prevalent during bull markets as opposed to bear markets. Additionally, the effect of sentiment on overtrading is observed to be more pronounced among individual investors in large-cap stocks compared to small- and mid-cap stocks.
Complexity vs Empirical Score
- Math Complexity: 3.0/10
- Empirical Rigor: 6.0/10 (capped from 8.5: no out-of-sample test; see penalty flags)
- Quadrant: Street Traders — empirical and applied, lower math complexity (effective empirical rigor capped from 8.5)
Why this score: The paper employs high-frequency data analysis and text mining for sentiment indices, demonstrating strong data/implementation focus, while the mathematics involved is standard econometric methods rather than dense advanced theory.
Research Flowchart
flowchart TD
A["Research Goal:<br>Effect of Investor Sentiment<br>on Intraday Overtrading"] --> B["Data Collection<br>High-Frequency Data: CSI 300 & 500<br>Eastmoney Guba Forum<br>01/2018 - 12/2022"]
B --> C["Methodology<br>Empirical Analysis & Regression<br>Trading Volume vs Sentiment Indices"]
C --> D["Key Findings"]
subgraph D ["Key Findings"]
F1["Sentiment significantly increases<br>intraday overtrading"]
F2["Effect is stronger for<br>Institutional Investors vs Individuals"]
F3["Effect is more pronounced<br>during Bull Markets"]
F4["Effect is stronger for Individuals<br>in Large-Cap Stocks"]
end
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 establish a metric for measuring investors' high‑frequency emotions and an indicator for overtrading behavior.
Source sentence
This study aims to establish a metric for measuring investors' high-frequency emotions and an indicator for overtrading behavior.
Assumes 5
- Investor sentiment measured in the previous hour influences the excess turnover rate in the current hour. model
Source sentence
In Formula (1), 𝐸𝑇𝑖,ℎ represents the excess turnover rate of stock 𝑖 during the trading day at Hour ℎ, and 𝑆𝑖,ℎ−1 represents the investor sentiment measure for stock 𝑖 during the time period from Hour ℎ−1 to Hour ℎ of the trading day.
- Data for the regression are collected starting from hour 2 of each trading day. data
Source sentence
the data for the regression model will be collected from hour 2 of each trading day onwards.
- The word‑dictionary approach to sentiment scoring is sufficiently accurate (≈80%). statistical
Source sentence
After comparing the text emotion calculated by the word dictionary approach with that obtained manually, it was found that the accuracy rate of the word dictionary approach is approximately 80.21%.
- Excess turnover ratio is calculated as the difference between current turnover ratio and its historical average. model
Source sentence
then calculates the excess turnover ratio by subtracting the current turnover ratio from the historical turnover ratio.
Show 1 more
- Transactions can be classified into large, medium, small, and ultra‑small orders to separate institutional and individual overtrading. model
Source sentence
This paper classifies each transaction into a large order, a medium-sized order, a small order, and an ultra‑small order based on the transaction amount and the number of shares traded.
Data
- Oriental Fortune Securities Forum posts text-news public
Source sentence
Through web scraping, we collected posts from the Oriental Fortune Securities Forum on the 1st of January, 2018, and the 30th of December, 2022.
- Intraday trading data market-prices commercial
Source sentence
We then obtained the intraday trading data of each stock from the Wind database by comparing the timestamps of the posts to the time period of the A-share market trading hours.
- CSI 300 and CSI 500 constituent lists other public
Source sentence
The CSI 500 Index, on the other hand, encompasses the 500 stocks from Shanghai and Shenzhen markets that are not included in the Hang Seng China
Costs
Not applicable
Finds 4
- Investor sentiment has a significantly positive impact on intraday overtrading, with coefficients ranging from 0.0673 to 0.1305 (p<0.01). statistical
Source sentence
Last hour sentiment 0.1305*** 0.0911*** 0.0673***
- The impact of sentiment on overtrading is stronger for institutional investors than for retail investors. statistical
Source sentence
Institutional investors are more susceptible to social media investor sentiment and are likely to engage in excessive trading behavior. This confirms Hypothesis H2a.
- Robustness tests (changing risk‑premium factor, market‑return ratio, transaction‑amount threshold) confirm the main results and support all research hypotheses. statistical
Source sentence
The results of the three robustness tests are consistent with the regression results in the empirical study, indicating a strong explanatory power for the model and support for all research hypotheses.
- Sentiment‑driven overtrading is more prevalent during bull markets than bear markets. statistical
Source sentence
sentiment-driven overtrading is found to be more prevalent during bull markets as opposed to bear markets.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 23 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.