Deep learning-based multi-dimensional investor sentiment and stock liquidity: Evidence from China

Zhiyi Wang; Jingru Guo; Gaoshan Wang; Xiaohong Shen; Zhiyi Wang; Jingru Guo; Gaoshan Wang; Xiaohong Shen

doi:10.3934/QFE.2025026

Quantitative Finance and Economics

2025, Volume 9, Issue 4: 745-779. doi: 10.3934/QFE.2025026

Previous Article Next Article

Research article

Deep learning-based multi-dimensional investor sentiment and stock liquidity: Evidence from China

1.
School of Management Science and Engineering, Shandong University of Finance and Economics, 7366 Erhuan Dong Road, Jinan 250014, China
2.
Institute of Digital Economy, Shandong University of Finance and Economics, 7366 Erhuan Dong Road, Jinan 250014, China
3.
School of Accounting, Shandong University of Finance and Economics, 7366 Erhuan Dong Road, Jinan 250014, China
4.
School of Computer Science and Technology, Shandong University of Finance and Economics, 7366 Erhuan Dong Road, Jinan 250014, China

Received: 21 April 2025 Revised: 04 August 2025 Accepted: 03 September 2025 Published: 11 October 2025
JEL Codes: G12, D53, C45, D84

Investor sentiment has long been recognized as a critical driver of financial market fluctuations and liquidity. However, conventional approaches that rely on aggregated sentiment measures often obscure the intrinsic multidimensional heterogeneity of sentiment, which results in information obfuscation and a reduced explanatory power. This study introduces a deep learning framework to overcome this limitation. By analyzing 11 million investor posts from the Chinese A-share market, we decompose the sentiment into six thematic dimensions: policy, psychological, social, climate, technological, and company. Our empirical models demonstrate that this multidimensional approach substantially enhances the explanation of liquidity fluctuations compared to a single aggregate index. We uncover significant heterogeneity in these effects. For blue-chip stocks, social, psychological, and company-related sentiments provide a persistent positive impact on liquidity. In contrast, technology stocks exhibit a greater sensitivity to a broader range of themes, with psychological, technological, and company sentiments driving more pronounced and nonlinear dynamics. These findings establish that the thematic source of sentiment is a critical determinant of its market impact, a linkage which traditional measures have masked. Our framework provides a more granular understanding of the sentiment-liquidity nexus, thus offering key insights for risk management and market surveillance.
- behavioral finance,
- Chinese market,
- deep learning,
- investor sentiment,
- sentiment analysis,
- liquidity
Citation: Zhiyi Wang, Jingru Guo, Gaoshan Wang, Xiaohong Shen. Deep learning-based multi-dimensional investor sentiment and stock liquidity: Evidence from China[J]. Quantitative Finance and Economics, 2025, 9(4): 745-779. doi: 10.3934/QFE.2025026

Related Papers:

Abstract

Investor sentiment has long been recognized as a critical driver of financial market fluctuations and liquidity. However, conventional approaches that rely on aggregated sentiment measures often obscure the intrinsic multidimensional heterogeneity of sentiment, which results in information obfuscation and a reduced explanatory power. This study introduces a deep learning framework to overcome this limitation. By analyzing 11 million investor posts from the Chinese A-share market, we decompose the sentiment into six thematic dimensions: policy, psychological, social, climate, technological, and company. Our empirical models demonstrate that this multidimensional approach substantially enhances the explanation of liquidity fluctuations compared to a single aggregate index. We uncover significant heterogeneity in these effects. For blue-chip stocks, social, psychological, and company-related sentiments provide a persistent positive impact on liquidity. In contrast, technology stocks exhibit a greater sensitivity to a broader range of themes, with psychological, technological, and company sentiments driving more pronounced and nonlinear dynamics. These findings establish that the thematic source of sentiment is a critical determinant of its market impact, a linkage which traditional measures have masked. Our framework provides a more granular understanding of the sentiment-liquidity nexus, thus offering key insights for risk management and market surveillance.

References

[1]	Amihud Y, Mendelson H (1986) Asset pricing and the bid-ask spread. J Financ Econ 17: 223–249. https://doi.org/10.1016/0304-405X(86)90065-6 doi: 10.1016/0304-405X(86)90065-6
[2]	Amihud, Y, Noh J (2021) Illiquidity and Stock Returns II: Cross-section and Time-series Effects. Rev Financ Stud 34: 2101–2123. https://doi.org/10.1093/rfs/hhaa080 doi: 10.1093/rfs/hhaa080
[3]	Ammari A, Chebbi K, Ben Arfa N (2023) How does the COVID-19 pandemic shape the relationship between Twitter sentiment and stock liquidity of US firms? Int Rev Financ Anal 88: 102633. https://doi.org/10.1016/j.irfa.2023.1026333 doi: 10.1016/j.irfa.2023.1026333
[4]	Antweiler W, Frank MZ (2004) Is all that talk just noise? The information content of internet stock message boards. J Financ 59: 1259–1294.https://doi.org/10.1111/j.1540-6261.2004.00662.x doi: 10.1111/j.1540-6261.2004.00662.x
[5]	Araci D (2019) Finbert: Financial sentiment analysis with pre-trained language models. arXiv preprint arXiv: 1908.10063. Available from: https://doi.org/10.48550/arXiv.1908.10063
[6]	Baker M, Stein JC (2004) Market liquidity as a sentiment indicator. J Financ Mark 7: 271–299. https://doi.org/10.1016/j.finmar.2003.11.005 doi: 10.1016/j.finmar.2003.11.005
[7]	Baker M, Wurgler J (2006) Investor sentiment and the cross‐section of stock returns. J Financ 61: 1645–1680. https://doi.org/10.1111/j.1540-6261.2006.00885.x doi: 10.1111/j.1540-6261.2006.00885.x
[8]	Baker M, Wurgler J (2007) Investor sentiment in the stock market. J Econ Perspect 21: 129–151. https://doi.org/10.1257/jep.21.2.129 doi: 10.1257/jep.21.2.129
[9]	Barberis N, Shleifer A, Vishny R (1998) A model of investor sentiment. J Financ Econ 49: 307–343. https://doi.org/10.1016/S0304-405X(98)00027-0 doi: 10.1016/S0304-405X(98)00027-0
[10]	Bernanke BS, Kuttner KN (2005) What explains the stock market's reaction to Federal Reserve policy? J Financ 60: 1221–1257. https://doi.org/10.1111/j.1540-6261.2005.00760.x doi: 10.1111/j.1540-6261.2005.00760.x
[11]	Blanchard O, Perotti R (2002) An empirical characterization of the dynamic effects of changes in government spending and taxes on output. Q J Econ 117: 1329–1368. https://doi.org/10.1162/003355302320935043 doi: 10.1162/003355302320935043
[12]	Bollen J, Mao H, Zeng X (2011) Twitter mood predicts the stock market. J Comput Sci 2: 1–8. https://doi.org/10.1016/j.jocs.2010.12.007 doi: 10.1016/j.jocs.2010.12.007
[13]	Brown GW, Cliff MT (2004) Investor sentiment and the near-term stock market. J Empir Financ 11: 1–27. https://doi.org/10.1016/j.jempfin.2002.12.001 doi: 10.1016/j.jempfin.2002.12.001
[14]	Brown GW, Cliff MT (2005) Investor sentiment and asset valuation. J Bus 78: 405–440. https://doi.org/10.1086/427633 doi: 10.1086/427633
[15]	Caldara D, Iacoviello M, Molligo P, et al. (2020) The economic effects of trade policy uncertainty. J Monetary Econ 109: 38–59. https://doi.org/10.1016/j.jmoneco.2019.11.002 doi: 10.1016/j.jmoneco.2019.11.002
[16]	Chiu J, Chung H, Ho KY, et al. (2018) Investor sentiment and evaporating liquidity during the financial crisis. Int Rev Econ Financ 55: 21–36. https://doi.org/10.1016/j.iref.2018.01.006 doi: 10.1016/j.iref.2018.01.006
[17]	Chordia T, Roll R, Subrahmanyam A (2001) Market liquidity and trading activity. J Financ 56: 501–530. https://doi.org/10.1111/0022-1082.00335 doi: 10.1111/0022-1082.00335
[18]	Da Z, Engelberg J, Gao P (2015) The sum of all FEARS investor sentiment and asset prices. Rev Financ Stud 28: 1–32. https://doi.org/10.1093/rfs/hhu072 doi: 10.1093/rfs/hhu072
[19]	Debata B, Dash SR, Mahakud J (2018) Investor sentiment and emerging stock market liquidity. Financ Res Lett 26: 15–31. https://doi.org/10.1016/j.frl.2017.11.006 doi: 10.1016/j.frl.2017.11.006
[20]	Devlin J, Chang MW, Lee K, et al. (2019) Bert: Pre-training of deep bidirectional transformers for language understanding. Paper presented at the Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies, volume 1 (long and short papers), 4171–4186, Minneapolis, Minnesota. Association for Computational Linguistics. Available from: https://doi.org/10.18653/v1/N19-1423.
[21]	Dunham LM, Garcia J (2021) Measuring the effect of investor sentiment on liquidity. Manag Financ 47: 59–85. https://doi.org/10.1108/MF-06-2019-0265 doi: 10.1108/MF-06-2019-0265
[22]	Fang J, Gozgor G, Lau CKM, et al. (2020) The impact of Baidu Index sentiment on the volatility of China's stock markets. Financ Res Lett 32: 101099. https://doi.org/10.1016/j.frl.2019.01.011 doi: 10.1016/j.frl.2019.01.011
[23]	Glosten LR, Milgrom PR (1985) Bid, ask and transaction prices in a specialist market with heterogeneously informed traders. J Financ Econ 14: 71–100. https://doi.org/10.1016/0304-405X(85)90044-3 doi: 10.1016/0304-405X(85)90044-3
[24]	Julio B, Yook Y (2012) Political uncertainty and corporate investment cycles. J Financ 67: 45–83. https://doi.org/10.1111/j.1540-6261.2011.01707.x doi: 10.1111/j.1540-6261.2011.01707.x
[25]	Kim SH, Kim D (2014) Investor sentiment from internet message postings and the predictability of stock returns. J Econ Behav Organ 107: 708–729. https://doi.org/10.1016/j.jebo.2014.04.015 doi: 10.1016/j.jebo.2014.04.015
[26]	Kyle AS (1985) Continuous auctions and insider trading. Econometrica: J Econometric Soc 1315–1335. https://doi.org/10.2307/1913210 doi: 10.2307/1913210
[27]	Li J, Ahn HJ (2024) Sensitivity of Chinese stock markets to individual investor sentiment: An analysis of Sina Weibo mood related to COVID-19. J Behav Exp Financ 41: 100860. https://doi.org/10.1016/j.jbef.2023.100860 doi: 10.1016/j.jbef.2023.100860
[28]	Liu JX, Leu JS, Holst S (2023) Stock price movement prediction based on stocktwits investor sentiment using FinBERT and ensemble SVM. Peerj Comput Sci 9: e1403. https://doi.org/10.7717/peerj-cs.1403 doi: 10.7717/peerj-cs.1403
[29]	Luo Y, Hong J, Ni C, et al. (2020) Can individual investors identify economic links? Evidence from initial quarterly earnings announcements within industries. China J Account Stud 8: 121–153. https://doi.org/10.1080/21697213.2020.1813970 doi: 10.1080/21697213.2020.1813970
[30]	Nyakurukwa K, Seetharam Y (2025) Investor sentiment networks: mapping connectedness in DJIA stocks. Financ Innovation 11: 4. https://doi.org/10.1186/s40854-024-00675-7 doi: 10.1186/s40854-024-00675-7
[31]	Pastor L, Veronesi P (2012) Uncertainty about government policy and stock prices. J Financ 67: 1219–1264. https://doi.org/10.1111/j.1540-6261.2012.01746.x doi: 10.1111/j.1540-6261.2012.01746.x
[32]	Ray P, Ganguli B, Chakrabarti A (2021) A Hybrid Approach of Bayesian Structural Time Series With LSTM to Identify the Influence of News Sentiment on Short-Term Forecasting of Stock Price. Ieee Trans Comput Social Syst 8: 1153–1162. https://doi.org/10.1109/tcss.2021.3073964 doi: 10.1109/tcss.2021.3073964
[33]	Simon HA (1955) A behavioral model of rational choice. Q J Econ, 99–118. https://doi.org/10.2307/1884852 doi: 10.2307/1884852
[34]	Souma W, Vodenska I, Aoyama H (2019) Enhanced news sentiment analysis using deep learning methods. J Comput Soc Sci 2: 33–46. https://doi.org/10.1007/s42001-019-00035-x doi: 10.1007/s42001-019-00035-x
[35]	Tetlock PC (2007) Giving content to investor sentiment: The role of media in the stock market. J Financ 62: 1139–1168. https://doi.org/10.1111/j.1540-6261.2007.01232.x doi: 10.1111/j.1540-6261.2007.01232.x
[36]	Wu DD, Zheng L, Olson DL (2014) A Decision Support Approach for Online Stock Forum Sentiment Analysis. Ieee Trans Syst Man Cybernetics-Syst 44: 1077–1087. https://doi.org/10.1109/tsmc.2013.2295353 doi: 10.1109/tsmc.2013.2295353
[37]	Zhang X, Li G, Li Y, et al. (2023) Which is more important in stock market forecasting: Attention or sentiment? Int Rev Financ Analy 89: 102732. https://doi.org/10.1016/j.irfa.2023.102732 doi: 10.1016/j.irfa.2023.102732
[38]	Zhang X, Lin L, Xiao M, et al. (2024) Do auditors care about what retail investors say? Evidence from China. Financ Res Lett 64: 105460. https://doi.org/10.1016/j.frl.2024.105460 doi: 10.1016/j.frl.2024.105460

QFE-09-04-026-s001.pdf

Reader Comments

Your name:*

Email:*
© 2025 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)