Please wait a minute...
Shopping Cart Email List Chinese
Advanced Search Fig/Tab
Frontiers of Engineering Management

ISSN 2095-7513

ISSN 2096-0255(Online)

CN 10-1205/N

Postal Subscription Code 80-905

Featured Articles  |  Most Downloaded  |  Most Reading
Online Submission Subscribe to Our RSS Content Feed
Front. Eng    2017, Vol. 4 Issue (2) : 201-211    https://doi.org/10.15302/J-FEM-2017018
RESEARCH ARTICLE
The influence of social media on stock volatility
Xianjiao WU, Xiaolin WANG(), Shudong MA, Qiang YE
School of Management, Harbin Institute of Technology, Harbin 150001, China
 Download: PDF(1692 KB)   HTML
 Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks
Abstract

This study explores the influence of social media on stock volatility and builds a feature model with an intelligence algorithm using social media data from Xueqiu.com in China, Sina Finance and Economics, Sina Microblog, and Oriental Fortune. We find that the effect of social factors, such as increased attention to a stock’s volatility, is more significant than public sentiment. A prediction model is introduced based on social factors and public sentiment to predict stock volatility. Our findings indicate that the influence of social media data on the next day’s volatility is more significant but declines over time.
Keywords stock volatility      social data      sentiment analysis      boosting algorithm     
Corresponding Author(s): Xiaolin WANG   
Just Accepted Date: 08 June 2017   Online First Date: 06 July 2017    Issue Date: 17 July 2017
 Cite this article:   
Xianjiao WU,Xiaolin WANG,Shudong MA, et al. The influence of social media on stock volatility[J]. Front. Eng, 2017, 4(2): 201-211.
 URL:  
https://academic.hep.com.cn/fem/EN/10.15302/J-FEM-2017018
https://academic.hep.com.cn/fem/EN/Y2017/V4/I2/201
Fig.1  Translated screen shot of Xueqiu.com — rankings data
Fig.2  Translated screen shot of Xueqiu.com — stock portfolio data
Fig.3  Translated screen shot of Xueqiu.com — real-time basic stock data
Stock_idSH600886SZ002702SZ002276SZ000025SZ002751
Company references125166741625659115593
Company references(standardized)10.540.50.470.45
Tab.1  Summary statistics of attention by company
VariablesDescription
Stock tickerUnique identification of stock
Share of stock in portfoliosShare of stock in top 100 portfolios of a category
Daily return of stock in portfoliosDaily return of stock shown in top 100 portfolios
Monthly return of stock in portfoliosMonthly return of stock shown in top 100 portfolios
Annual return of stock in portfoliosAnnual return of stock shown in top 100 portfolios
Total yield in portfoliosTotal yield of stock shown in top 100 portfolios
Investment style of portfoliosInvestment style of stock shown in top 100 portfolios
Attention of stock in portfoliosAttention of stock in portfolios
DateDate when occurred
Tab.2  Stock portfolio variable descriptions after refactoring
AttributeDescription
Segment of an essay“In the first week after the Spring Festival, the A-shares began to rebound as expected, and some specific stocks had excessive growth within a few days. In the wake of the rebound, the prices of A-shares fluctuate slightly on Friday. The Shanghai Composite Index closed at 2860.02 with a decline of 0.10%, while the Growth Enterprise Index rose by 0.92%. The stocks of shipbuilding, electronic information industry, the water and gas supplier, the electronic components, the foreign trade industry, the glass industry, the food industry, and the aircraft industry rose more than stocks of petroleum industry, the real estate, the steel industry, and the brewing industry.”
Emotional word extractionRebound, growth, rebound, fluctuate, decline, rose
Emotional polarityPositive
Tab.3  An example of emotional polarity assessment
Fig.4  Sketch of sentiment classification training effect
VariablesDescriptionCorrelationSignificance
Attention increases this weekAttention increases in the stock this week0.619**0.0021
Attention is hottestAttention of the stock in the hottest list0.418**0.0049
Discussion increases this weekDiscussion increases on the stock this week0.406**0.0052
Discussion is hottestDiscussion of the stock in the hottest list0.469**0.0040
Transaction sharing increases in this weekTransaction sharing increases on the stock this week0.364**0.0089
Transaction sharing is hottestThe amount of transaction sharing of the stock in the hottest list0.547**0.0037
Tab.4  The correlational analysis between social data and volume
Fig.5  Daily chart of the Shanghai Composite Index
Fig.6  The importance of features
Fig.7  The importance of social attributes
Fig.8  Conceptual model
ClassPrecisionRecallF1-scoreSupport
Decline0.770.770.7722983
Rise0.830.830.8330644
Avg/total0.810.810.8153627
Tab.5  The precision, recall rate, F1, and support of social data
Fig.9  Training social data P/R curve
Predicting date2.182.192.222.232.242.252.26
Accuracy0.8410.6490.9430.6130.5860.1640.675
Predicting date2.293.13.23.33.43.73.8
Accuracy0.8080.8050.8390.5770.6780.8490.621
Tab.6  Predicting accuracy of stock volatility in t+1
Datet+1t+2t+3
Cross validation accuracy0.670.520.47
Random sampling accuracy0.860.850.82
Tab.7  The predicting effect in t+1, t+2, and t+3
1 Antweiler W, Frank M Z (2004). Is all that talk just noise? The information content of Internet stock message boards. Journal of Finance, 59(3): 1259–1294
2 Asur S, Huberman B A (2010). Predicting the future with social media. 2010 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT). IEEE, 1: 492–499
3 Bollen J, Mao H, Zeng X (2016). Twitter mood predicts the stock market. Eprint Arxiv, 2(1): 1–8
4 Chen Y (2016). Predicting stock trading volume through social media data. 2016-04-01, 
5 Cheng W Y, Lin J (2013). The relationship between stock index and investor sentiment in social media. Management Science, 26(5): 111–119
6 Choi J J, Laibson D, Metrick A (2000). Does the Internet increase trading? Evidence from investor behavior in 401(k) plans. Ssrn Electronic Journal, 64(12): 10–11
7 Choudhury M D, Sundaram H, John A, Seligmann D D (2008). Can blog communication dynamics be correlated with stock market activity? Hypertext 2008, Proceedings of the ACM Conference on Hypertext and Hypermedia, Pittsburgh, PA, USA, 55–60
8 Das S R, Chen M Y (2007). Yahoo! for Amazon: sentiment extraction from  small  talk  on  the  web.  Management  Science,  53(9):  1375–1388
9 Fang L, Peress J (2008). Media coverage and the cross-section of stock returns. Social Science Electronic Publishing, 64(5): 2023–2052
10 Feng L N (2013). A Study on Influence of Open Source Information Flow on Stock Volatility. Dissertation for Master’s Degree. Tianjin: Tianjin University
11 Freedman S, Jin G Z (2011). Learning by doing with asymmetric information: evidence from prosper. com.Nber Working Papers,2011: 203–212
12 Gao X P (2009). The infectivity of emotion in social network. Pictorial of Science, (9): 20–22
13 Geurts P, Ernst D, Wehenkel L (2006). Extremely randomized trees. Machine Learning, 63(1): 3–42
14 He W, Guo L, Shen J, Akula V (2016). Social media-based forecasting: a case study of tweets and stock prices in the financial services industry. Journal of Organizational and End User Computing, 28(2): 74–91
15 Li Y L, Li S C, Yang G H (2007). The correlation between stock market volume and price. Journal of Hebei University of Economy and Trade, 28(2): 65–70
16 Liu Y, Lv B F, Peng G (2011). Predictive power of internet search data for stock market: a theoretical analysis and empirical test. Economic Management, 33(1): 172–180
17 Oliveira N, Cortez P, Areal N (2017). The impact of microblogging data for stock market prediction: using Twitter to predict returns, volatility, trading volume and survey sentiment indices. Expert Systems with Applications, 73: 125–144
18 Pak A, Paroubek P (2010). Twitter as a corpus for sentiment analysis and opinion mining. In: Proceedings of the International Conference on Language Resources and Evaluation. Valletta, Malta
19 Pang L, Li S S, Zhang H (2012). Research on stock investor sentiment tendency based on microblog. Computer Science, B06: 249–252
20 Salton G, Mcgill M J (1983). Introduction to modern information retrieval. Library Management, 32(4/5): 373–374
21 Schumaker R P, Chen H (2009a). Textual analysis of stock market prediction using breaking financial news: the AZFin text system. ACM Transactions on Information Systems, 27(2): 1–19 
22 Schumaker R P, Chen H (2009b). A quantitative stock prediction system based on financial news. Information Processing & Management, 45(5): 571–583
23 Shi M J (2005). An analysis of volume’s impact on stock yield. Statistics & Information Forum, 20(2): 60–62
24 Sivic J, Zisserman A (2009). Efficient visual search of videos cast as text retrieval. IEEE Transactions on Pattern Analysis and Machine Intelligence, 31(4): 591–606
25 Tetlock P C (2007). Giving content to investor sentiment: the role  of media in the stock market. Journal of Finance, 62(3): 1139–1168
26 Tetlock P C, Saar-Tsechansky M, Macskassy S (2008). More than words: quantifying language to measure firms’ fundamentals. Journal of Finance, 63(3): 1437–1467 
27 Tirunillai S, Tellis G J (2012). Does chatter really matter? Dynamics of user-generated content and stock performance. Marketing Science, 31(2): 198–215
28 Tumarkin R, Whitelaw R F (2001). News or noise? Internet postings and stock prices. Prehospital and Disaster Medicine, 57(3): 41–51
29 Wen F H, Xiao J L, Huang C X (2014). Research on influence of investors sentiment on stock price. Journal of Management Science, 17(3): 60–69
30 Yang D J, Yang J, Zhan X J (2014). Feature selection algorithm based on the random forest. Journal of Jilin University: Engineering Science, 44(1): 137–141
31 Yang X, Lv B F, Peng G (2013). The influence of emergency on stock market: an analysis based on online searching. Practice and Knowledge of Mathematics, 43(23): 17–28
32 Zou K H, O’Malley A J, Mauri L (2007). Receiver-operating characteristic analysis for evaluating diagnostic tests and predictive models. Circulation, 115(5): 654–657
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed