基于朴素贝叶斯法的投资者情绪度量及其对股票特质风险的影响

doi:10.16381/j.cnki.issn1003-207x.2021.0028

Abstract

Abstract:

With the vigorous development of web2.0 and the rapid development of text mining technology， different subjects create and share content on social media. In the process of information dissemination and sharing， the information structure of the stock market has been changed， which further strengthens the sentiment tendency of investors and ultimately affects their decisions. Investors driven by sentiment will induce noise trading and cause changes in stock prices， which is manifested as changes in stock idiosyncratic risk. However， the influence of investor sentiment on idiosyncratic risk is rarely mentioned. Therefore， by crawling the real-time post content of the individual stock bar of Oriental Fortune.com， the daily investor sentiment of individual stock is constructed using the Naive Bayesian method. A panel regression model is used to investigate the influence of investor sentiment on idiosyncratic risk.The research is supplemented on the impact of investor sentiment and the influencing factors of idiosyncratic risk， which is helpful for a deeper understanding of the influence mechanism of investor sentiment on stock risk from the perspective of a high frequency， and has certain guidance and reference value for investors， listed companies and regulators.The results show that　（1） Investor sentiment in the lagging period and the current period all have a significant positive impact on idiosyncratic risk（IR）， indicating that idiosyncratic risk increases when investor sentiment tends to be optimistic. （2） The effect of investor sentiment on idiosyncratic risk varies with the change of the difficulty of stock valuation. Stocks with a lower book-to-market and analyst tracking number are more difficult to value， and investor sentiment has a greater impact on idiosyncratic risk. （3） Considering the effect of arbitrage restriction， with the increase in the degree of short-selling restrictions， the effect of investor sentiment on idiosyncratic risk is significantly enhanced. After a series of robust tests， the above conclusions are still robust.

Key words: investor sentiment, idiosyncratic risk, text mining, arbitrage restriction

CLC Number:

F830.91

Haiyuan Yin,Wenjuan Kou. Investor Sentiment Based on Naive Bayes Method and Its Impact on Stock Idiosyncratic Risk[J]. Chinese Journal of Management Science, 2024, 32(4): 38-47.

Figures/Tables 9

指标

α 0

α 1

隔夜收益率

-0.974^***

(-431.20)

0.543^***

(3.48）

动量指标

-0.980^***

(-435.89）

0.669^***

(28.65)

投资者情绪

-0.890^***

(-402.74)

0.370^***

(112.77)

References 26

1	Renault T. Intraday online investor sentiment and return patterns in the U.S. stock market［J］. Journal of Banking & Finance， 2017， 84（11）：25-40.
2	Ballinari D， Behrendt S. Structural breaks in online investor sentiment： a note on the nonstationarity of financial chatter［J］.Finance Research Letters， 2020， 35（11）：101479.
3	黄创霞，温石刚，杨鑫，等.个体投资者情绪与股票价格行为的互动关系研究［J］.中国管理科学， 2020， 28（3）： 191-200.
	Huang C X， Wen S G， Yang X， et al. The interactive relationship between individual investor sentiment and stock price behaviors［J］. Chinese Journal of Management Science， 2020， 28（3）： 191-200.
4	尹海员，吴兴颖. 投资者高频情绪对股票日内收益率的预测作用［J］. 中国工业经济， 2019（8）： 80-98.
	Yin H Y， Wu X Y. Predictive effect of high-frequency investor sentiment on the intraday stocks return［J］.China Industrial Economics，2019（8）：80-98.
5	Tsukioka Y， Yasutomo J， Yanagi T. Investor sentiment extracted from internet stock message boards and IPO puzzles［J］. International Review of Economics and Finance， 2018， 56（3）：205-217.
6	杨晓兰，沈翰彬，祝宇. 本地偏好、投资者情绪与股票收益率：来自网络论坛的经验证据［J］. 金融研究，2016（12）： 143-158.
	Yang X L， Shen H B， Zhu Y.The effect of local bias in investor attention and investor sentiment on stock markets： evidence from online forum［J］. Journal of Financial Research，2016（12）：143-158.
7	Antweiler W， Frank M Z. Is all that talk just noise？ theinformation content of internet stock message boards［J］. Journal of Finance， 2004， 59（3）：1259-1294.
8	Kim S H， Kim D.Investor sentiment from internet message postings and the predictability of stock returns［J］.Journal of Economic Behavior & Organization， 2014， 107（11）：708-729.
9	Leung H， Ton T. The impact of internet stock message boards on cross-sectional returns of small-capitalization stocks［J］.Journal of Banking & Finance， 2015， 55（6）：37-55.
10	Li Y L， Bu H， Li J H， et al. The role of text-extracted investor sentiment in Chinese stock price prediction with the enhancement of deep learning［J］.International Journal of Forecasting， 2020， 36（4）：1541-1562.
11	Campbell J Y， Lettau M， Xu M Y. Have individual stocks become more volatile？ An empirical exploration of idiosyncratic risk［J］. Journal of Finance， 2001， 56（1）：1-43.
12	Ang A， Hodrick R J， Xing Y， et al.The cross-section of volatility and expected returns［J］.Journal of Finance， 2006， 61（6）： 259-300.
13	Gu M， Kang W J， Xu B. Limits of arbitrage and idiosyncratic volatility： evidence from China stock market ［J］.Journal of Banking & Finance，2018， 86（4）：240-258.
14	Fenner R G， Han Y F， Huang Z D. Idiosyncratic volatility shocks， behavior bias， and cross-sectional stock returns［J］. The Quarterly Review of Economics and Finance， 2020， 75（8）：276-293.
15	王立章，王咏梅，王志诚. 控制权、现金流权与股价同步性［J］. 金融研究， 2016（5）：97-110.
	Wang L Z， Wang Y M， Wang Z C. Stock price synchronicity and the separation of ownership between control rights［J］. Journal of Financial Research，2016（5）：97-110.
16	陈作华，刘子旭. 政企关系与企业特质风险［J］. 管理科学， 2019， 32（4）：48-61.
	Chen Z H， Liu Z X. Government-enterprise relationship and firm idiosyncratic risk［J］. Journal of Management Science，2019， 32（4）：48-61.
17	Caglayan M O， Xue W J， Zhang L W. Global investigation on the country-level idiosyncratic volatility and its determinants［J］. Journal of Empirical Finance， 2020， 55：143-160.
18	熊熊，许克维，沈德华.投资者情绪与期货市场功能——基于沪深300股指期货的研究［J］. 系统工程理论与实践， 2020， 40（9）： 2252-2268.
	Xiong X， Xu K W， Shen D H. Investor sentiment and futures market functions： evidence from CSI300 index futures market［J］. Systems Engineering-Theory & Practice，2020，40（9）： 2252-2268.
19	陆昌，刘洋，杨晓光. 投资者情绪的不对称性及其原因——来自中国市场的实证［J］. 系统科学与数学， 2020，40（4）： 612-633.
	Lu C， Liu Y， Yang X G. Asymmetry of investor sentiment and the reason： evidence from Chinese stock market［J］. Journal of Systems Science and Mathematical Sciences，2020，40（4）： 612-633.
20	DeLong J B， Shleifer A， Summers L H， et al.Noise trader risk in financial markets［J］. Journal of Political Economy， 1990， 98（4）：703-738.
21	王健，李明操，郭文轩，等. 个股情绪、定向增发与大股东利益输送——来自中国A股上市公司的经验证据［J］. 中国管理科学， 2022，30（9）：23-35.
	Wang J， Li M C， Guo W X， et al. Individual stock sentiment， private placement， and tunneling： empirical evidence from Chinese A-share listed companies［J］.Chinese Journal of Management Science， 2022，30（9）：23-35.
22	Polk C， Sapienza P. The stock market and corporate investment： a test of catering theory［J］. Review of Financial Studies， 2009， 22（1）：187-217.
23	Fu F. Idiosyncratic risk and the cross-section of expected stock returns［J］. Journal of Financial Economics， 2009， 91（1）：24-37.
24	Gao Z Y， Ren H H， Zhang B H.Googlinginvestor sentiment around the world［J］. Journal of Financial and Quantitative Analysis， 2020， 55（2）：549-580.
25	武佳薇，汪昌云，陈紫琳，等.中国个人投资者处置效应研究——一个非理性信念的视角［J］.金融研究，2020（2）： 147-166.
	Wu J W， Wang C Y， Chen Z L， et al. A study of disposition effect among China’s individual investors： the perspective of irrational beliefs［J］. Journal of Financial Research，2020（2）：147-166.
26	Liu S M. Investor sentiment and stock market liquidity ［J］.Journal of Behavioral Finance，2015， 16（1）：51-67.

变量分类	变量名称	变量符号	变量说明
被解释变量	特质风险	IR	由式(3)和式(5)计算得出
解释变量	投资者情绪	Sentiment	由式(1)计算得出
控制变量	资产收益率	ROA	净利润与总资产的比值
	资产负债率	Lev	负债总额与资产总额的比值
	公司规模	Size	日个股流通市值的自然对数
	换手率	Turn	日个股交易股数与流通股份的比值
	上市年限	Age	IPO年度到计算特质风险年度为止的年数
	流动比例	CRR	流动资产与总资产的比值
	经营性现金净流量	Cfo	经营活动产生的现金净流量与总资产的比值
	机构持股比例	Institution	机构投资者持股总数与总股数的比值

变量	样本量	均值	标准差	最小值	中位数	最大值	偏度	峰度
IR_it	72268	0.050	0.047	0.010	0.035	0.285	2.669	11.384
Sentiment_it	72268	-0.976	0.591	-2.485	-0.981	0.575	0.005	3.051
ROA	72268	0.018	0.039	-0.166	0.015	0.156	-0.709	10.606
Lev	72268	0.514	0.206	0.098	0.519	0.960	-0.129	2.294
Turn	72268	1.807	2.387	0.079	0.919	13.688	2.791	11.849
Age	72268	18.491	5.638	2.000	19.000	28.000	-0.845	3.402
Size	72268	16.172	1.021	14.419	16.043	19.495	0.767	3.572
CRR	72268	0.505	0.216	0.065	0.513	0.936	-0.113	2.161
Cfo	72268	0.019	0.057	-0.158	0.016	0.186	0.014	4.208
Institution	72268	53.013	20.741	0.006	53.562	120.107	-0.130	2.533
BM	72268	0.812	0.261	0.110	0.853	1.375	-0.562	2.709
个股日均发帖量	72268	46.132	83.741	1	25	4406	12.027	331.735
单条帖子阅读量	72268	12279.806	7084.525	175	13919	22506	-0.263	1.649

变量	IR	Sentiment	ROA	Lev	Turn	Age	Size	CRR	Cfo	Institution
IR	1	0.130^***	-0.163^***	0.012^***	0.611^***	-0.070^***	-0.062^***	-0.134^***	-0.095^***	-0.097^***
Sentiment	0.106^***	1	0.008^***	-0.003^**	-0.018^***	0.018^***	0.017^***	0.001	-0.001	0.026^***
ROA	-0.137^***	0.017^***	1	-0.245^***	-0.250^***	-0.052^***	0.339^***	-0.079^***	0.481^***	0.188^***
Lev	0.008^**	-0.006	-0.217^***	1	-0.066^***	0.016^***	0.086^***	0.188^***	-0.046^***	0.152^***
Turn	0.680^***	0.021^***	-0.191^***	-0.024^***	1	0.114^***	-0.323^***	0.159^***	-0.176^***	-0.043^***
Age	-0.056^***	0.019^***	-0.026^***	-0.005	0.094^***	1	-0.207^***	0.083^***	-0.129^***	-0.139^***
Size	-0.084^***	0.020^***	0.330^***	0.070^***	-0.228^***	-0.240^***	1	-0.070^***	0.150^***	0.542^***
CRR	-0.100^***	-0.002	-0.054^***	0.170^***	0.124^***	0.090^***	-0.100^***	1	-0.293^***	-0.071^***
Cfo	-0.107^***	0.005	0.489^***	-0.045^***	-0.168^***	-0.092^***	0.189^***	-0.247^***	1	0.129^***
Institution	-0.111^***	0.029^***	0.178^***	0.138^***	-0.278^***	-0.196^***	0.546^***	-0.095^***	0.129^***	1

变量	(1)	(2)	(3)
变量	IR	IR	IR
Sentiment_it-1	0.171^***		0.102^***
Sentiment_it-1	(3.320)		(4.471)
Sentiment_it		0.212^***	0.192^***
Sentiment_it		(6.451)	(7.795)
ROA	-0.007^***	-0.008^***	-0.008^***
ROA	(-3.301)	(-3.357)	(-3.332)
Lev	0.005^***	0.004^**	0.004^**
Lev	(2.210)	(2.188)	(2.194)
Turn	0.013^***	0.013^***	0.013^***
Turn	(35.64)	(35.84)	(35.70)
Age	-0.010^**	-0.010^**	-0.010^**
Age	(-1.743)	(-1.672)	(-1.563)
Size	-0.048^***	-0.041^***	-0.043^***
Size	(-10.18)	(10.21)	(-10.20)
CRR	-0.047^**	-0.046^**	-0.046^**
CRR	(-2.111)	(-2.098)	(-2.098)
Cfo	-0.010^***	-0.010^***	-0.010^***
Cfo	(-4.621)	(-3.639)	(-3.639)
Institution	-0.007^**	-0.007^**	-0.006^**
Institution	(-2.683)	(-2.700)	(-2.694)
截距项	-0.477^**	-0.476^**	-0.476^**
截距项	(-2.741)	(-2.741)	(-2.741)
行业/年度	控制	控制	控制
聚类层面	公司	公司	公司
Adj-R²	0.446	0.446	0.487
F值	163.323	161.058	165.798

变量	Panel A：账面市值比分组回归结果				Panel B：分析师人数分组回归结果
	(1) 低BM	(2) 高BM	(3) 低BM	(4) 高BM	(1) 分析师人数低	(2) 分析师人数高	(3) 分析师人数低	(4) 分析师人数高
	IR	IR	IR	IR	IR	IR	IR	IR
Sentiment_it-1	0.161^***	0.101	0.091^***	0.083	0. 102^***	0.008^**	0.104^***	0.008^**
Sentiment_it-1	(3.508)	(0.404)	(4.366)	(1.278)	(2.587)	(2.568)	(3.376)	(2.525)
Sentiment_it			0.281^***	0.109^*			0.122^***	0.013^**
Sentiment_it			(5.464)	(1.762)			(5.925)	(2.800)
ROA	-0.019^***	-0.006^***	-0.032^***	-0.006^***	-0.014^**	-0.098^**	-0.013^**	-0.098^**
ROA	(-4.079)	(-4.229)	(-3.118)	(-3.206)	-(2.451)	(-2.730)	(-2.444)	(-2.741)
Lev	0.009^**	0.002^**	0.008^**	0.007^**	0.016^**	0.004^**	0.016^**	0.004^**
Lev	(2.311)	(2.072)	(2.292)	(2.068)	(2.385)	(2.158)	(2.390)	(2.173)
Turn	0.018^***	0.017^***	0.013^***	0.017^***	0.013^***	0.016^***	0.013^***	0.014^***
Turn	(30.52)	(20.02)	(30.65)	(20.06)	(31.591)	(15.665)	(31.600)	(15.711)
Age	-0.012^**	-0.005^**	-0.012^**	-0.005^**	-0.013^**	-0.006^**	-0.014^**	-0.006^**
Age	(-1.930)	(-2.513)	(-1.940)	(-2.543)	(-2.043)	(-2.020)	(-1.950)	(-1.962)
Size	-0.047^***	-0.041^***	-0.046^***	-0.042^***	-0.050^***	-0.041^***	-0.049^***	-0.040^***
Size	(-8.793)	(-6.466)	(-8.796)	(-6.478)	(-9.218)	(-5.409)	(-9.171)	(-5.426)
CRR	-0.012^**	-0.101^***	-0.011^**	-0.099^***	-0.048^*	-0.031^**	-0.047^*	-0.032^*
CRR	(-2.313)	(-2.490)	(-2.311)	(-3.456)	(-1.555)	(-1.806)	(-1.528)	(-1.590)
Cfo	-0.008^***	-0.018^***	-0.007^***	-0.019^***	-0.012^***	-0.012^***	-0.010^***	-0.012^***
Cfo	(-3.331)	(-4.132)	(-5.321)	(-4.160)	(-6.141)	(-5.556)	(-8.340)	(-4.357)
Institution	-0.003^**	-0.007^***	-0.003^**	-0.007^***	-0.004^**	-0.006^*	-0.005^**	-0.006^*
Institution	(-1.912)	(-2.480)	(-1.933)	(-2.428)	(-2.342)	(-1.760)	(-2.355)	(-1.761)
截距项	-0.487^**	-0.542^**	-0.484^**	-0.542^***	-0.475^**	-0.773^***	-0.471^**	-0.772^***
截距项	(-1.924)	(-2.820)	(-1.914)	(-2.853)	(-1.987)	(-4.795)	(-1.970)	(-4.783)
行业/年度	控制	控制	控制	控制	控制	控制	控制	控制
聚类层面	公司	公司	公司	公司	公司	公司	公司	公司
Adj-R²	0.454	0.427	0.454	0.428	0.485	0.312	0.485	0.313
F值	125.167	51.143	136.691	50.034	141.308	137.809	143.498	139.432

Investor Sentiment Based on Naive Bayes Method and Its Impact on Stock Idiosyncratic Risk

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 9

References 26

Related Articles 15

Metrics

Comments

Recommended 0

变量	(1) 类型1	(2) 类型2	(3) 类型1	(4) 类型2
变量	IR	IR	IR	IR
Sentiment_it-1	0.118^***	0.062^**	0.114^***	0.091^**
Sentiment_it-1	(3.581)	(2.000)	(3.414)	(2.80)
Sentiment_it			0.132^***	0.090^***
Sentiment_it			(5.542)	(5.786)
ROA	-0.014^**	-0.020^**	-0.015^**	-0.019^**
ROA	(-2.580)	(-2.459)	(-2.591)	(-2.455)
Lev	0.019^***	0.007^***	0.018^***	0.011^**
Lev	(7.219)	(5.213)	(4.340)	(3.252)
Turn	0.012^***	0.015^***	0.013^***	0.015^***
Turn	(30.57)	(18.06)	(30.66)	(18.06)
Age	-0.010^**	-0.002^*	-0.011^**	-0.003^*
Age	(-1.978)	(-1.442)	(-1.789)	(-1.534)
Size	-0.048^***	-0.039^**	-0.048^***	-0.040^***
Size	(-9.186)	(-6.256)	(-9.101)	(-6.209)
CRR	-0.063^**	-0.047^*	-0.062^**	-0.047^*
CRR	(-2.392)	(-1.518)	(-2.365)	(-1.419)
Cfo	-0.016^***	-0.028^**	-0.015^***	-0.029^**
Cfo	(-5.601)	(-2.415)	(-8.519)	(-2.402)
Institution	-0.008^**	-0.004^*	-0.007^**	-0.003^*
Institution	(-2.433)	(-1.766)	(-2.451)	(-1.750)
截距项	-0.483^*	-0.583^***	-0.481^*	-0.584^***
截距项	(-1.791)	(-3.011)	(-1.798)	(-3.110)
行业/年度	控制	控制	控制	控制
聚类层面	公司	公司	公司	公司
Adj-R²	0.476	0.397	0.476	0.398
F值	129.742	43.571	135.219	44.439

变量	(1)	(2)
	工具变量法
	IR	IR
Sentiment_it-1	0.135^***	0.389^***
Sentiment_it-1	(10.700)	(8.289)
Sentiment_it		0.130^***
Sentiment_it		(6.771)
ROA	-0.048^***	-0.102^***
ROA	(-3.221)	(-4.507)
Lev	0.039^***	0.070^***
Lev	(3.134)	(3.232)
Turn	0.010^***	0.008^***
Turn	(17.326)	(8.259)
Age	-0.008^**	-0.009^**
Age	(-2.004)	(-2.553)
Size	-0.043^***	-0.042^***
Size	(-6.913)	(-4.545)
CRR	-0.060^**	-0.070^*
CRR	(-1.844)	(-1.335)
Cfo	-0.008^***	-0.026^***
Cfo	(-3.334)	(-3.619)
Institution	-0.001^*	-0.001^**
Institution	(1.645)	(2.009)
行业/年度	控制	控制
聚类层面	公司	公司
Adj-R²	0.344	0.356
F值	115.475	54.868

[1]	WANG Dan-yang, YAO Lu-shi. Mutual Funds’ Dynamic Liquidity Preference-Based on Expected Market Volatility and Investor Sentiment [J]. Chinese Journal of Management Science, 2023, 31(5): 39-48.
[2]	YIN Hai-yuan, WU Xing-ying. Investor Sentiment, Order Flow Imbalance and Stock Liquidity [J]. Chinese Journal of Management Science, 2023, 31(5): 60-70.
[3]	LI Cheng-gang, JIA Hong-ye, ZHAO Guang-hui, FU Hong. Credit Risk Warning of Listed Companies Based on Information Disclosure Text:Empirical Evidence from Management Discussion and Analysis of the Chinese Annual Report [J]. Chinese Journal of Management Science, 2023, 31(2): 18-29.
[4]	Jiang-ze DU,Xi-zhuo CHEN,Le-an YU. Can Executives with IT Background Promote FinTech Innovation? [J]. Chinese Journal of Management Science, 2023, 31(12): 69-78.
[5]	Ming-yu LI,Dong-xiao NIU,Zheng-sen JI,Bo-wen SHI,Xin-yi LAN,Huan-fen ZHANG. Research on Multi-value Chain Management Risk Identification and Control of Electric Power Manufacturing Industry Oriented to the Construction of Data Space System [J]. Chinese Journal of Management Science, 2023, 31(11): 349-360.
[6]	WU Xi, WANG Yu-dong. The Effects of Air Pollution on Stock Prices of Heavy-air-pollution Enterprises: A Policy Intervention Perspective on Economic Value [J]. Chinese Journal of Management Science, 2023, 31(1): 267-276.
[7]	WU Jin-yan, WANG Peng. Factors Affecting the Risk Contagion of the Stock Market: An Evidence from Industry-Level Data [J]. Chinese Journal of Management Science, 2022, 30(8): 57-68.
[8]	ZHU Li, LIU Xiang-li, YANG Xiao-guang. Does Investor Sentiment Affect the Price Dynamic Relationship of Stock Index Futures-spot Market? [J]. Chinese Journal of Management Science, 2022, 30(4): 52-62.
[9]	WANG Yu-hang, DANG Yan-zhong, XU Zhao-guang. Mining Automobile Quality Problems Based on the Characteristics of Forum Data [J]. Chinese Journal of Management Science, 2021, 29(9): 201-212.
[10]	CHEN Wen-bo, CHEN Lang-nan. Reactions of Stock Investors to Earnings Announcements——A Perspective from Gambling Preference [J]. Chinese Journal of Management Science, 2021, 29(9): 1-11.
[11]	XIANG Cheng, YANG Jun. Who Gambles in the Market? A Study on Mutual Funds’ Preferences for Lottery-like Stocks [J]. Chinese Journal of Management Science, 2021, 29(11): 224-236.
[12]	ZHANG Bo, HU Wen-xiu, YANG Xi-an. Research on the Formatting Mechanism of Investor Sentiment [J]. Chinese Journal of Management Science, 2021, 29(1): 185-195.
[13]	KONG Ji-hong, YUE Wei. Is Dynamic Nelson-Siegel Model Compatible with Arbitrage Free——Empirical Evidence from China Treasury Yield Curves [J]. Chinese Journal of Management Science, 2020, 28(4): 61-72.
[14]	HUANG Chuang-xia, WEN Shi-gang, YANG Xin, WEN Feng-hua, YANG Xiao-guang. The Interactive Relationship between Individual Investor Sentiment and Stock Price Behaviors [J]. Chinese Journal of Management Science, 2020, 28(3): 191-200.
[15]	WANG Wei, HE Ling, KEVIN Zhu, SUN Rui, WANG Hong-wei. The Periodic Impact of Linguistic Cues to Update Signalson Successful Crowdfunding Campaigns Among Categories [J]. Chinese Journal of Management Science, 2020, 28(11): 155-166.