Please wait a minute...
Advanced Search
数据分析与知识发现  2019, Vol. 3 Issue (9): 98-114    DOI: 10.11925/infotech.2096-3467.2018.1223
     研究论文 本期目录 | 过刊浏览 | 高级检索 |
财经媒介信息对股票市场的影响研究: 基于情感分析的实证 *
岑咏华1(),谭志浩1,吴承尧2
1 南京理工大学经济管理学院 南京 210094
2 南京农业大学金融学院 南京 210095
Impacts of Financial Media Information on Stock Market: An Empirical Study of Sentiment Analysis
Yonghua Cen1(),Zhihao Tan1,Chengyao Wu2
1 School of Economics and Management, Nanjing University of Science and Technology, Nanjing 210094, China
2 College of Finance, Nanjing Agricultural University, Nanjing 210095, China
全文: PDF(1994 KB)   HTML ( 20
输出: BibTeX | EndNote (RIS)      
摘要 

【目的】考察媒介信息所蕴含的情感信号对于股票市场的影响效应。【方法】利用LSTM深度神经网络方法对主流财经网站的新闻、股吧、博客文本的情感信息进行识别和提取, 构建自回归分布滞后模型和面板回归模型, 从宏观市场以及微观股票资产两个层面实证揭示财经媒介信息所蕴含的情感对股票市场表现的关联影响。【结果】(1) 媒介信息情感的倾向性变化在短期内导致价格的显著同向变化, 同时受投资者追涨杀跌驱动, 市场成交量显著提升。更长时间上, 市场对于媒介信息情感的过度反应将反转, 价格回归。(2) 媒介信息情感波动和分歧程度与价格负相关, 与成交量呈现非线性的U形关系。(3) 投资者对于积极情感的反应更及时更强烈, 理性调整更缓慢, 而对于消极情感呈现出显著的处置效应。(4) 相较于单一的利好或利空情绪, 投资者对于市场意见高分歧、利空与利好同在的反应更强烈, 股价在投资者过度交易中持续下落。【局限】针对有着不同语法和语义表达特征的不同媒介信息类型, 未选择不同的模型和参数进行词向量编码, 情感分析的准确度可能受到影响。【结论】本研究可为媒介信息效应以及投资者情绪影响相关研究提供新的视角和洞见, 对金融监管具有理论、方法和实践指导意义。

服务
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章
岑咏华
谭志浩
吴承尧
关键词 财经媒介信息情感分析投资者情绪股票市场情感倾向    
Abstract

[Objective] This paper aims to study the impacts of media coverage on stock market. [Methods] We used the LSTM deep neural networks to evaluate the sentiments of the online news, forum posts and blogs from leading financial websites. Then, we established autoregressive distributed lag model and panel regression model to test the relationship between media information sentiments and stock market performance from the perspectives of macro market and individual stocks. [Results] (I) In the short term, the positive and negative sentiments significantly changed the stock prices and led to overreaction. In the longer term, the stock market reversed. (II) There were a negative relationship between sentiment volatility/discrepancy and stock prices, and a U-shaped nonlinear correlation between sentiment discrepancy and trading. (III) Investors reacted more immediately and strongly to positive sentiments, and the rational correction of this overreaction was slower than those of the negative information. (IV) High discrepancy of sentiments led to more over-trading than high consensus. [Limitations] The accuracy of sentiment analysis needs to be improved with more complex models. [Conclusions] Our research provides theoretical, methodological and practical implications for financial supervision and regulation.

Key wordsFinancial Media Information    Sentiment Analysis    Investor Sentiment    Stock Market    Sentiment Orientation
收稿日期: 2018-11-04     
中图分类号:  F832.51 G35  
基金资助:*本文系国家自然科学基金项目“投资者有限关注与证券市场监管: 基于大数据和计算实验的方法”(项目编号: 71503130);国家自然科学基金项目“社会化影响下个体信息认知处理中的扭曲与偏见机制研究”(项目编号: 71471089);国家自然科学基金项目“突发事件网民负面情感的模型检测研究”(项目编号: 71774084)
引用本文:   
岑咏华,谭志浩,吴承尧. 财经媒介信息对股票市场的影响研究: 基于情感分析的实证 *[J]. 数据分析与知识发现, 2019, 3(9): 98-114.
Yonghua Cen,Zhihao Tan,Chengyao Wu. Impacts of Financial Media Information on Stock Market: An Empirical Study of Sentiment Analysis. Data Analysis and Knowledge Discovery, DOI:10.11925/infotech.2096-3467.2018.1223.
链接本文:  
http://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/10.11925/infotech.2096-3467.2018.1223
图1  整体研究框架
信息粒度 信息来源 信息类型分布(条)
新闻 博客 股吧
宏观市场 新浪财经 25 490 21 426 2 347
东方财富 37 044 1 735 8 615
网易财经 13 254 - -
合计 75 788 23 161 10 962
个股 新浪财经 14 087 - 1 272
东方财富 20 977 - 6 217
网易财经 2 758 - -
合计 37 822 - 7 489
表1  财经媒介信息的粒度、来源和类型分布
图2  基于LSTM的情感分析过程
算法 Accuracy Precision Recall F1-score
SVM 0.5862 0.5910 0.9673 0.7337
BP 0.5903 0.5898 0.9897 0.7420
CNN 0.7525 0.8115 0.7556 0.7825
LSTM 0.8144 0.8109 0.8933 0.8501
表2  不同情感分析模型评估
数据面板 变量 平均值 最大值 最小值 标准差 偏度 峰度 单整阶数 PP检验值
Panel.A: 媒介
信息情感
SO 0.6744 0.7174 0.6214 0.0202 -0.3347 2.5678 (0) -9.8203***
SV 0.0805 0.1024 0.0628 0.0087 0.2056 2.6996 (0) -10.7562***
SO*(日内) 0.2606 18.1554 -0.6484 0.7985 10.5351 138.0403 (0) 9375.8602***
SV*(日内) 0.0168 0.5842 0.0000 0.0397 4.1882 25.9605 (0) 9644.6624***
SO*(日间) 1.1709 56.1808 -2.2263 3.0606 8.5154 95.7669 (0) 5086.3990***
SV*(日间) 0.0183 0.5218 0.0000 0.0396 3.9059 22.7119 (0) 6677.4028***
Panel.B: 上证
指数
M_Clpr 3271.5939 3559.4700 3066.8000 123.4484 0.4051 2.5142 (1) -9.3615***
M_Trdvol 1.6774E+10 2.8055E+10 1.0032E+10 4.0519E+09 0.9380 3.0704 (0) -4.7527***
Panel.C: 沪深
300指数
M_Clpr 4016.4328 4389.8900 3748.6400 157.1514 0.3873 2.5773 (1) -9.6843***
M_Trdvol 1.1402E+10 2.1982E+10 5.9030E+09 3.9969E+09 1.0704 3.2561 (0) -4.6446***
Panel.D: 沪深
300成分股日
内数据
S_Price 26.9833 788.4200 0.0000 63.8346 9.7353 105.9154 (0) 493.9201***
S_Turnrate 0.8038 52.2100 0.0000 1.4678 11.0528 237.6628 (0) 9230.6062***
S_Trdvol 3.9976E+05 2.0554E+07 0.0000 7.9409E+05 7.1529 90.5288 (0) 9301.6712***
Panel.E: 沪深
300成分股日
间数据
S_Price 26.2359 788.4200 0.0000 59.7262 10.1848 117.1104 (0) 571.3529***
S_Turnrate 1.1299 54.6326 0.0000 1.6606 9.4680 200.1871 (0) 2234.6131***
S_Trdvol 5.7799E+07 2.0554E+09 0.0000 1.0438E+08 6.3401 66.3050 (0) 2263.0469***
表3  描述性统计
SO SV M_Clpr ΔM_Clpr M_Trdvol ΔM_Trdvol
SV -0.8058***(-14.2729) 1
M_Clpr 0.2053**(2.2001) -0.1559(-1.6550) 1
ΔM_Clpr 0.3686***(4.1591) -0.3621***(-4.0748) 0.1464(1.5518) 1
M_Trdvol 0.1644*(1.7480) -0.0152(-0.1595) 0.6099***(8.0727) -0.1828*(-1.9496) 1
ΔM_Trdvol -0.0174(-0.1830) 0.0444(0.4659) 0.0807(0.8488) 0.0002(0.0025) 0.3612***(4.0631) 1
表4  市场变量间相关性分析
时间窗口 模型变量 SO* SV* S_Price S_Turnrate S_Trdvol
日内 SV* 0.3303***(64.6454) 1
S_Price 0.0058(1.0866) -0.0200***(-3.7089) 1
S_Turnrate 0.0218***(4.0363) 0.0367***(6.7842) 0.0791***(14.6636) 1
S_Trdvol 0.0397***(7.3400) 0.0049(0.9117) 0.7873***(235.9581) 0.1847***(34.7206) 1
日间 SV* 0.4181***(58.2939) 1
S_Price 0.0279***(3.5399) -0.0315***(-4.0028) 1
S_Turnrate 0.0291***(3.6974) 0.0283***(3.5920) 0.0210***(2.6712) 1
S_Trdvol 0.1452***(18.5834) 0.0663***(8.4207) -0.4386***(-61.8087) 0.3056***(40.6518) 1
表5  股票变量间相关性分析
图3  市场层面情感倾向性与情感波动性趋势
图4  股票层面情感倾向性与情感波动性日内(上)和日间(下)趋势图(以平安银行为例)
模型参数 上证指数 沪深300指数
ΔM_Clpr M_Trdvol ΔM_Trdvol ΔM_Clpr M_Trdvol ΔM_Trdvol
β1 -0.0393(-0.3694) 0.4290***(4.7764) -0.4176***(-4.7963) 0.1110(1.0399) 0.5181***(5.7020) -0.3854***(-4.4091)
β2 0.0710(0.6783) 0.2929***(3.3671) 0.0455(0.4151) 0.2988***(3.3333)
β3 0.2319**(2.3682) 0.0792(0.7286)
β4 -0.3160***(-3.0866)
α0 0.2601***(3.9104) 0.7823(1.0547) 0.1923(0.2469) 0.1792***(3.0040) 0.7821(0.8209) 0.5083(0.5180)
α1 -0.1425**(-1.9684) 0.1936(0.3977) -0.1027(-0.1992) -0.0575(-1.2537) -0.2551(-0.4001) -0.4175(-0.6351)
α2 -0.0545(-0.7292) 0.4420(0.8957) 0.1652(0.3155) -0.0903**(-2.0177) -0.0796(-0.1228) -0.2281(-0.3410)
α3 1.5275**(2.1995) 0.9961(1.3636) 0.0809(1.2780) 1.3085(1.4561) 1.0764(1.1618)
φ1 -0.0216(-0.2936) -3.3924**(-2.2590) -1.6521(-1.0725) -0.0371(-0.4945) -2.9089*(-1.6240) -1.9466(-1.0681)
φ2 0.0008(0.3214) 0.0243**(2.0119) 0.0106(0.8507) 0.0026(1.1941) 0.0212(1.3318) 0.0063(0.4024)
φ3 -0.3908**(-2.2786) 5.1211***(2.6306) 4.9407**(2.3693) -0.0693(-0.3881) 6.4401**(2.1208) 7.3834**(2.3627)
Adj-R2 0.2222 0.6425 0.1980 0.1760 0.6760 0.1991
F检验 3.8890*** 25.4259*** 4.7128*** 2.9235*** 29.3452*** 4.6261***
DW检验 1.9281 2.0727 2.0971 1.9995 2.1309 2.1738
表6  模型I的检验结果
模型参数 ΔM_Clpr M_Trdvol ΔM_Trdvol
pos neg pos neg pos neg
β -0.3715**(-2.1792) -0.0120(-0.0608) 0.4490***(3.1025) -0.3555**(-2.5437) -0.2765(-0.3310) -0.4349**(-2.7380)
α 0.2860**(2.3767) 0.4950**(1.9885) 7.4977**(2.0899) -4.8370**(-2.3099) 1.7185(0.6219) -5.2740*(-1.7577)
φ1 0.5611(0.7474) -0.3896**(-2.2654) 1.2498(0.7533) 5.9920***(5.2644) 1.9406(1.3206) 2.3985**(2.8297)
φ2 0.0253(0.9429) 0.0132*(1.7407) 0.0200(0.5485) -0.3297***(-4.5713) -0.0708*(-1.6820) 0.0011(0.0412)
φ3 -0.7123**(-2.8195) -0.2178(-0.6097) -4.5734(-0.5401) -8.0594*(-1.7879) 9.3053(1.5347) 0.6143(0.1347)
Adj-R2 0.5683 0.1337 0.2341 0.5635 0.0671 0.1852
F检验 4.9503*** 2.2955** 2.3169** 9.8227*** 1.8157 2.8643**
DW检验 1.9056 2.1264 1.9352 1.9176 1.9295 2.0456
表7  模型I-1和模型I-2的检验结果(τ=0)
模型参数 ΔM_Clpr M_Trdvol ΔM_Trdvol
pos neg pos neg pos neg
β -0.2501**(-1.9908) -0.1919*(-1.7017) 0.1242(1.0936) 0.0615(0.3420) -0.2795(-1.5251) -0.3661**(-2.2737)
α 0.0324(0.1551) -0.3477**(-1.9886) 0.4697(0.2429) 5.6319**(1.9980) -2.3744(-1.0554) 5.6451**(2.0743)
φ1 -0.0342(-0.3788) 0.1821(0.9595) 0.6900(0.5667) 4.1655**(2.1409) -0.0701(-0.0611) 0.6854(0.7876)
φ2 0.0082*(1.9181) 0.0018(0.7772) 0.1332**(2.5089) -0.0952(-1.6272) 0.0131(0.2918) 0.0219(1.1469)
φ3 -0.0012(-0.0035) 0.4750(1.1750) 2.8001(0.9078) -4.9336(-0.9411) 3.4674(0.9560) 3.3466(0.9068)
Adj-R2 0.0513 0.0937 0.6303 0.4493 0.0955 0.3261
F检验 3.6631** 1.9827* 12.3680*** 6.4412*** 1.7865 4.1466***
DW检验 2.1342 1.9298 1.9921 1.9636 2.0749 2.1658
表8  模型I-1和模型I-2的检验结果(τ=1)
系数 日内时间窗口 日间时间窗口
S_Price S_Turnrate S_Trdvol S_Price S_Turnrate S_Trdvol
β1 1.0419***(3.1008) 0.4836***(7.4929) 0.1763***(3.9575) 0.8195***(29.7544) 0.2790***(5.0571) 0.4380***(14.2150)
β2 -0.0659(-0.1733) -0.0143(-0.2211) 0.1563***(3.5235) 0.1008***(7.1111) 0.0450(0.9260) 0.1388***(4.3277)
β3 -0.1308(-1.4000) 0.3078***(5.2342) 0.2116***(5.8476) -0.0856***(-7.1874) 0.0610(1.3234) 0.0983***(3.4314)
β4 0.1460***(3.8939) -0.1769***(-2.9079) 0.1499***(4.9389) 0.0586(1.3990) 0.0177(0.7614)
α0 -0.0079(-1.1581) 0.0428***(3.0398) -0.0156(-0.3134) 0.0013***(2.7776) 0.0628***(4.7443) 0.0118***(3.9790)
α1 0.0049(1.3370) 0.0174**(2.5475) 0.0433**(2.4730) 0.0002(0.7938) 0.0305***(4.3534) 0.0053***(3.2730)
α2 0.0108***(2.6223) 0.0018(0.1972) 0.0402**(2.1566) -0.0004(-1.0476) 0.0074(1.1787) 0.0012(0.6941)
α3 0.0100(1.3951) -0.0039(-0.4924) -0.0249(-0.8146) -0.0008*(-1.8529) -0.0063(-1.0986) -0.0003(-0.1673)
α4 -0.0008**(-2.2191) -0.0108**(-1.9966) 0.0005(0.3162)
α5 -0.0006**(-2.4154) -0.0061(-0.5134) 0.0038(1.2891)
φ1 0.0134(1.4299) -0.2761(-1.2856) -0.0515(-0.5463) 0.0886***(3.2119) 0.9190***(3.7292) 0.2893***(5.9758)
φ2 -0.0003(-0.9376) 0.0001(0.6940) 0.0002(0.8675) 4.30E-05(1.3502) 0.0004*(1.8060) 0.0001***(2.7287)
φ3 -0.0304(-1.0054) -1.4728***(-3.6531) -1.3399**(-2.0467) -0.0591***(-4.0935) 0.3259(0.7185) -0.3159(-1.4846)
Adj-R2 0.9734 0.6551 0.4026 0.9988 0.8263 0.9060
F检验 1933.2790*** 120.7757*** 37.0884*** 10371.630*** 71.9508*** 143.8802***
DW检验 2.2439 2.0858 2.0994 1.8643 1.8949 1.9971
表9  模型II的检验结果
模型参数 上证指数 沪深300指数
ΔM_Clpr M_Trdvol ΔM_Trdvol ΔM_Clpr M_Trdvol ΔM_Trdvol
β1 -0.1102(-1.1785) 0.4581***(5.0539) -0.4239***(-4.8674) 0.1013(0.8891) 0.5201***(5.6764) -0.3939***(-4.4891)
β2 0.0413(0.3070) 0.3237***(3.6520) 0.0047(0.0413) 0.3163***(3.4943)
β3 0.1861**(1.9714) 0.0529(0.4657)
β4 -0.2666**(-2.4748)
α0 -0.5198**(-2.5972) -0.1105(-0.0676) 0.5121(0.3011) -0.3553**(-2.5293) 0.7364(0.3496) 1.1212(0.5193)
α1 0.2040(1.6034) -0.7024(-0.6307) -0.6070(-0.5209) 0.1647(1.1172) -0.4665(-0.3224) -0.5340(-0.3592)
α2 -0.0161(-0.1344) -1.2836(-1.1469) -1.2593(-1.0750) -0.0370(-0.2509) -0.8808(-0.6021) -1.0682(-0.7116)
α3 -1.8543(-1.1809) -1.4447(-0.8817) -0.5064(-0.2489) -0.4815(-0.2303)
φ1 -0.0391(-0.4622) -2.5279*(-1.7014) -1.3678(-0.9055) -0.0358(-0.4690) -2.0237(-1.1382) -1.2186-0.6774
φ2 0.0015(0.4909) 0.0177(1.4483) 0.0083(0.6681) 0.0030(1.3391) 0.0196(1.2001) 0.00560.3512
φ3 -0.4147***(-3.1097) 5.1425**(2.5719) 5.0155**(2.3968) -0.0967(-0.5256) 6.5306**(2.1243) 7.3684**2.3455
Adj-R2 0.1573 0.6235 0.1897 0.1197 0.6673 0.1807
F检验 2.8887*** 23.5626*** 4.6459*** 2.2242** 28.3223*** 4.4350***
DW检验 1.9112 2.0625 2.1030 2.1344 2.1228 2.1722
表10  模型III的检验结果
模型参数 ΔM_Clpr M_Trdvol ΔM_Trdvol
disc cons disc cons disc cons
β -0.2538*(-1.6962) -0.6805***(-5.0723) -0.2219(-1.2873) -0.0275(-0.1691) -1.2525***(-4.5781) -0.2457**(-2.0928)
α -0.0115***(-3.3752) -0.0070**(-2.4173) 0.1453**(2.4297) -0.0743**(-2.1615) 0.0948*(1.6927) 0.0123(0.3605)
φ1 -0.2593**(-1.9888) 0.8614***(3.9183) 1.8352(0.9914) -1.4771(-0.6751) 9.9305**(2.2148) 0.2866(0.2362)
φ2 0.0080(1.6079) 0.0176**(2.2726) -0.0031(-0.0390) 0.1161(1.7832) -0.4636*(-1.7377) -0.0092(-0.2434)
φ3 -0.2633(-1.0137) -0.7711(-1.0434) 5.9980(1.3167) 1.3742(0.4788) 0.5949(0.0645) 5.1284*(1.6774)
Adj-R2 0.2373 0.2910 0.2140 0.2587 0.5105 0.0804
F检验 3.6138** 3.3227** 2.4614* 2.7453** 3.4336** 1.9746*
DW检验 2.1564 1.9097 1.9608 2.1733 2.0836 2.1967
表11  模型III-1和模型III-2的检验结果(τ=0)
模型参数 日内时间窗口 日间时间窗口
S_Price S_Turnrate S_Trdvol S_Price S_Turnrate S_Trdvol
β1 0.9758***(127.9304) 0.4708***(55.2554) 0.2084***(23.6092) 0.9845***(157.4324) 0.4707***(57.3960) 0.4627***(25.4147)
β2 0.0105(1.0808) 0.0088(0.9611) 0.1656***(18.2184) 0.0081(0.9326) 0.0321***(3.5926) 0.1392***(5.4947)
β3 -0.1416***(-16.1052) 0.2662***(31.1766) 0.2222***(25.2909) -0.0211**(-2.4326) 0.1660***(18.5293) 0.0881***(3.8828)
β4 0.1453***(22.3432) -0.1478***(-18.8960) 0.1428***(15.8484) -0.0082(-1.3292) 0.0193**(2.3537) 0.0566***(3.6867)
α0 -0.2280***(-3.8408) 0.4975**(2.4455) -0.6131(-1.2443) -0.0044(-1.0736) 0.7090***(4.0463) 0.2852***(3.0739)
α1 -0.0534(-1.3081) -0.2204(-1.5752) -1.1163***(-3.2943) -0.0050**(-2.1971) 0.2936***(2.9746) 0.1147**(2.1422)
α2 0.0762*(1.8167) -0.2455*(-1.7066) -0.5292(-1.5191) -0.0051**(-2.4429) -0.0028(-0.0317) -0.0065(-0.1438)
α3 0.1609***(2.6952) 0.4222**(2.0629) 1.1480**(2.3164) -0.0047*(-1.9280) -0.1805*(-1.7136) -0.0786(-1.5652)
α4 -0.0037(-1.5844) -0.2392**(-2.3738) -0.1016**(-2.1267)
α5 -0.0021(-1.3809) -0.1790***(-2.6887) -0.0753**(-2.3853)
φ1 0.0083(0.4348) -0.2895***(-4.3808) 0.0692(0.4361) 0.4615***(105.2908) -0.2785***(-4.6535) 0.2543***(5.3329)
φ2 -0.0003***(-6.8114) 0.0001(0.5784) -0.0007**(-1.9755) -0.0001(-0.6924) -0.0001(-0.6070) -0.0001(-0.5329)
φ4 -0.0320(-0.6411) -1.3355***(-7.6351) -0.6839*(-1.6714) -0.0189***(-4.7210) -0.7591***(-4.6478) -0.4113**(-2.2898)
Adj-R2 0.9692 0.6480 0.4360 0.9893 0.6739 0.8827
F检验 1925.0570*** 113.5124*** 48.2358*** 10280.110*** 136.7012*** 489.0216***
DW检验 1.9374 2.0624 2.0832 1.9109 2.0228 1.9674
表12  模型IV的检验结果
[1] Grossman S J, Stiglitz J E . On the Impossibility of Informationally Efficient Markets[J]. The American Economic Review, 1980,70(3):393-408.
[2] De Long 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.
[3] Daniel K, Hirshleifer D, Subrahmanyam A . Investor Psychology and Security Market Under- and Overreactions[J]. The Journal of Finance, 1998,53(6):1839-1885.
[4] Baker M, Wurgler J . Investor Sentiment in the Stock Market[J]. Journal of Economic Perspectives, 2007,21(2):129-152.
[5] Drakos K. Terrorism Activity, Investor Sentiment , Stock Returns[J]. Review of Financial Economics, 2010,19(3):128-135.
[6] Edmans A, Garcia D, Norli Ø . Sports Sentiment and Stock Returns[J]. The Journal of Finance, 2007,62(4):1967-1998.
[7] Goetzmann W N, Kim D, Kumar A , et al. Weather-Induced Mood, Institutional Investors, and Stock Returns[J]. The Review of Financial Studies, 2014,28(1):73-111.
[8] Huang D, Jiang F, Tu J , et al. Investor Sentiment Aligned: A Powerful Predictor of Stock Returns[J]. The Review of Financial Studies, 2015,28(3):791-837.
[9] Hirshleifer D, Shumway T . Good Day Sunshine: Stock Returns and the Weather[J]. The Journal of Finance, 2003,58(3):1009-1032.
[10] Kamstra M J, Kramer L A, Levi M D . Winter Blues: A SAD Stock Market Cycle[J]. American Economic Review, 2003,93(1):324-343.
[11] Kaplanski G, Levy H . Sentiment and Stock Prices: The Case of Aviation Disasters[J]. Journal of Financial Economics, 2010,95(2):174-201.
[12] Kaplanski G, Levy H, Veld C , et al. Do Happy People Make Optimistic Investors?[J]. Journal of Financial and Quantitative Analysis, 2015,50(1-2):145-168.
[13] Lo A W, Repin D V, Steenbarger B N . Fear and Greed in Financial Markets: A Clinical Study of Day-Traders[J]. American Economic Review, 2005,95(2):352-359.
[14] Tetlock P C . Giving Content to Investor Sentiment: The Role of Media in the Stock Market[J]. The Journal of Finance, 2007,62(3):1139-1168.
[15] 刘晓星, 张旭, 顾笑贤 , 等. 投资者行为如何影响股票市场流动性?——基于投资者情绪、信息认知和卖空约束的分析[J]. 管理科学学报, 2016,19(10):87-100.
( Liu Xiaoxing, Zhang Xu, Gu Xiaoxian , et al. How Does Investor Behavior Affect Stock Market Liquidity? Analysis of Investor Sentiment, Information Cognition and Short-Sale Constraints[J]. Journal of Management Sciences in China, 2016,19(10):87-100.)
[16] 许海川, 周炜星 . 情绪指数与市场收益: 纳入中国波指 (iVX) 的分析[J]. 管理科学学报, 2018,21(1):88-96.
( Xu Haichuan, Zhou Weixing . Sentiment Index and Market Return Considering the iVX[J]. Journal of Management Sciences in China, 2018,21(1):88-96.)
[17] Brown G W, Cliff M T . Investor Sentiment and Asset Valuation[J]. The Journal of Business, 2005,78(2):405-440.
[18] Lee C M C, Shleifer A, Thaler R H . Investor Sentiment and the Closed-End Fund Puzzle[J]. The Journal of Finance, 1991,46(1):75-109.
[19] Scheinkman J A, Xiong W . Overconfidence and Speculative Bubbles[J]. Journal of Political Economy, 2003,111(6):1183-1220.
[20] Baker M, Wurgler J . A Catering Theory of Dividends[J]. The Journal of Finance, 2004,59(3):1125-1165.
[21] Whaley R E . The Investor Fear Gauge[J]. The Journal of Portfolio Management, 2000,26(3):12-17.
[22] Frazzini A, Lamont O A . Dumb Money: Mutual Fund Flows and the Cross-Section of Stock Returns[J]. Journal of Financial Economics, 2008,88(2):299-322.
[23] Baker M, Wurgler J . Investor Sentiment and the Cross-Section of Stock Returns[J]. The Journal of Finance, 2006,61(4):1645-1680.
[24] Stambaugh R F, Yu J, Yuan Y . The Short of It: Investor Sentiment and Anomalies[J]. Journal of Financial Economics, 2012,104(2):288-302.
doi: 10.1016/j.jfineco.2011.12.001
[25] 高大良, 刘志峰, 杨晓光 . 投资者情绪, 平均相关性与股市收益[J]. 中国管理科学, 2015,23(2):10-20.
( Gao Daliang, Liu Zhifeng, Yang Xiaoguang. Investor Sentiment , Average Correlation and Stock Market Return[J]. Chinese Journal of Management Science, 2015,23(2):10-20.)
[26] 宋泽芳, 李元 . 投资者情绪与股票特征关系[J]. 系统工程理论与实践, 2012,32(1):27-33.
( Song Zefang, Li Yuan . Relation Between Investor Sentiment and Stock Characteristics[J]. Systems Engineering-Theory & Practice, 2012,32(1):27-33.)
[27] Ben-Rephael A, Kandel S, Wohl A . Measuring Investor Sentiment with Mutual Fund Flows[J]. Journal of Financial Economics, 2012,104(2):363-382.
doi: 10.1016/j.jfineco.2010.08.018
[28] 易志高, 茅宁 . 中国股市投资者情绪测量研究: CICSI的构建[J]. 金融研究, 2009(11):174-184.
( Yi Zhigao, Mao Ning . Research on the Measurement of Investor Sentiment in Chinese Stock Market: The CICSI’s Construction[J]. Journal of Financial Research, 2009(11):174-184.)
[29] Da Z, Engelberg J, Gao P . The Sum of All FEARS Investor Sentiment and Asset Prices[J]. The Review of Financial Studies, 2015,28(1):1-32.
[30] Loughran T , McDonald B. Textual Analysis in Accounting and Finance: A Survey[J]. Journal of Accounting Research, 2016,54(4):1187-1230.
[31] Kearney C, Liu S . Textual Sentiment in Finance: A Survey of Methods and Models[J]. International Review of Financial Analysis, 2014,33:171-185.
[32] Loughran T, McDonald B . When is a Liability Not a Liability? Textual Analysis, Dictionaries, and 10-Ks[J]. The Journal of Finance, 2011,66(1):35-65.
[33] Kraus M, Feuerriegel S . Decision Support from Financial Disclosures with Deep Neural Networks and Transfer Learning[J]. Decision Support Systems, 2017,104:38-48.
[34] Xing F Z, Cambria E, Welsch R E . Natural Language Based Financial Forecasting: A Survey[J]. Artificial Intelligence Review, 2018,50(1):49-73.
[35] 苏治, 卢曼, 李德轩 . 深度学习的金融实证应用: 动态, 贡献与展望[J]. 金融研究, 2018(5):111-126.
( Su Zhi, Lu Man, Li Dexuan . Deep Learning in Financial Empirical Applications: Dynamics, Contributions and Prospects[J]. Journal of Financial Research, 2018(5):111-126.)
[36] Schumaker R P, Chen H . Textual Analysis of Stock Market Prediction Using Breaking Financial News: The AZFin Text System[J]. ACM Transactions on Information Systems, 2009, 27(2): Article No. 12.
[37] Sun L, Najand M, Shen J . Stock Return Predictability and Investor Sentiment: A High-Frequency Perspective[J]. Journal of Banking & Finance, 2016,73:147-164.
[38] Bartov E, Faurel L, Mohanram P S . Can Twitter Help Predict Firm-Level Earnings and Stock Returns?[J]. The Accounting Review, 2017,93(3):25-57.
[39] Chan S W K, Chong M W C . Sentiment Analysis in Financial Texts[J]. Decision Support Systems, 2017,94:53-64.
[40] Checkley M S, Higón D A, Alles H . The Hasty Wisdom of the Mob: How Market Sentiment Predicts Stock Market Behavior[J]. Expert Systems with Applications, 2017,77:256-263.
[41] Chen H, De P, Hu Y J , et al. Wisdom of Crowds: The Value of Stock Opinions Transmitted Through Social Media[J]. The Review of Financial Studies, 2014,27(5):1367-1403.
[42] Deng S, Huang Z J, Sinha A P , et al. The Interaction Between Microblog Sentiment and Stock Return: An Empirical Examination[J]. MIS Quarterly, 2018,42(3):895-918.
[43] Kelly S, Ahmad K . Estimating the Impact of Domain-Specific News Sentiment on Financial Assets[J]. Knowledge-Based Systems, 2018,150:116-126.
[44] Li T, van Dalen J, van Rees P J . More than Just Noise? Examining the Information Content of Stock Microblogs on Financial Markets[J]. Journal of Information Technology, 2018,33(1):50-69.
[45] Siganos A, Vagenas-Nanos E, Verwijmeren P . Facebook’s Daily Sentiment and International Stock Markets[J]. Journal of Economic Behavior & Organization, 2014,107:730-743.
[46] Wu D D, Zheng L, Olson D L . A Decision Support Approach for Online Stock Forum Sentiment Analysis[J]. IEEE Transactions of Systems, Man, and Cybernetics: Systems, 2014,44(8):1077-1087.
[47] Zhang X, Zhang Y, Wang S , et al. Improving Stock Market Prediction via Heterogeneous Information Fusion[J]. Knowledge-Based Systems, 2018,143:236-247.
[48] 石善冲, 朱颖楠, 赵志刚 , 等. 基于微信文本挖掘的投资者情绪与股票市场表现[J]. 系统工程理论与实践, 2018,38(6):1404-1412.
( Shi Shanchong, Zhu Yingnan, Zhao Zhigang , et al. The Investor Sentiment Mined from WeChat Text and Stock Market Performance[J]. Systems Engineering-Theory & Practice, 2018,38(6):1404-1412.)
[49] 汪昌云, 武佳薇 . 媒体语气, 投资者情绪与IPO定价[J]. 金融研究, 2015(9):174-189.
( Wang Changyun, Wu Jiawei. Media Tone , Investor Sentiment and IPO Pricing[J]. Journal of Financial Research, 2015(9):174-189.)
[50] Antweiler W, Frank M Z . Is All That Talk Just Noise? The Information Content of Internet Stock Message Boards[J]. The Journal of Finance, 2004,59(3):1259-1294.
[51] Behrendt S, Schmidt A . The Twitter Myth Revisited: Intraday Investor Sentiment, Twitter Activity and Individual-Level Stock Return Volatility[J]. Journal of Banking & Finance, 2018,96:355-367.
[52] Das S R, Chen M Y . Yahoo! for Amazon: Sentiment Extraction from Small Talk on the Web[J]. Management Science, 2007,53(9):1375-1388.
[53] Kim S H, Kim D . Investor Sentiment from Internet Message Postings and the Predictability of Stock Returns[J]. Journal of Economic Behavior & Organization, 2014,107:708-729.
[54] Strauß N, Vliegenthart R, Verhoeven P . Intraday News Trading: The Reciprocal Relationships Between the Stock Market and Economic News[J]. Communication Research, 2018,45(7):1054-1077.
[55] Nassirtoussi A K, Aghabozorgi S, Wah T Y , et al. Text Mining for Market Prediction: A Systematic Review[J]. Expert Systems with Applications, 2014,41(16):7653-7670.
doi: 10.1016/j.eswa.2014.06.009
[56] Johnman M, Vanstone B J, Gepp A . Predicting FTSE 100 Returns and Volatility Using Sentiment Analysis[J]. Accounting & Finance, 2018,58(S1):253-274.
[57] Li B, Chan K C C, Ou C , et al. Discovering Public Sentiment in Social Media for Predicting Stock Movement of Publicly Listed Companies[J]. Information Systems, 2017,69:81-92.
[58] Li Q, Wang T, Li P , et al. The Effect of News and Public Mood on Stock Movements[J]. Information Sciences, 2014,278:826-840.
[59] Nguyen T H, Shirai K, Velcin J . Sentiment Analysis on Social Media for Stock Movement Prediction[J]. Expert Systems with Applications, 2015,42(24):9603-9611.
[60] 部慧, 解峥, 李佳鸿 , 等. 基于股评的投资者情绪对股票市场的影响[J]. 管理科学学报, 2018,21(4):86-101.
( Bu Hui, Xie Zheng, Li Jiahong , et al. Investor Sentiment Extracted from Internet Stock Message Boards and Its Effect on Chinese Stock Market[J]. Journal of Management Sciences in China, 2018,21(4):86-101.)
[61] 岑咏华, 张灿, 吴承尧 , 等. 互联网加剧投资者有限理性研究综述[J]. 外国经济与管理, 2018,40(6):129-140.
( Cen Yonghua, Zhang Can, Wu Chengyao , et al. How Does Internet Escalate Investor Bounded Rationality? A Literature Review[J]. Foreign Economics & Management, 2018,40(6):129-140.)
[62] Mikolov T, Chen K, Corrado G , et al. Efficient Estimation of Word Representations in Vector Space[OL]. arXiv Preprint, arXiv:1301.3781.
[63] Mikolov T, Sutskever I, Chen K, et al. Distributed Representations of Words and Phrases and Their Compositionality [C]// Proceedings of the 2013 Neural Information Processing Systems Conference. 2013: 3111-3119.
[64] Lai S, Liu K, He S , et al. How to Generate a Good Word Embedding[J]. IEEE Intelligent Systems, 2016,31(6):5-14.
[65] Rumelhart D E, Hinton G E, Williams R J . Learning Representations by Back-Propagating Errors[J]. Nature, 1986,323(6088):533-536.
[66] LeCun Y, Bottou L, Bengio Y , et al. Gradient-Based Learning Applied to Document Recognition[J]. Proceedings of the IEEE, 1998,86(11):2278-2324.
[67] Hopfield J J . Neural Networks and Physical Systems with Emergent Collective Computational Abilities[J]. Proceedings of the National Academy of Sciences, 1982,79(8):2554-2558.
[68] Pollack J B . Recursive Distributed Representations[J]. Artificial Intelligence, 1990,46(1-2):77-105.
[69] Young T, Hazarika D, Poria S , et al. Recent Trends in Deep Learning Based Natural Language Processing[J]. IEEE Computational Intelligence Magazine, 2018,13(3):55-75.
[70] Hochreiter S, Bengio Y, Frasconi P , et al. Gradient Flow in Recurrent Nets: The Difficulty of Learning Long-Term Dependencies[A]//Kolen J F, Kremer S C. A Field Guide to Dynamical Recurrent Networks[M]. Wiley-IEEE Press, 2001: 237-243.
[71] Hochreiter S, Schmidhuber J . Long Short-Term Memory[J]. Neural Computation, 1997,9(8):1735-1780.
[72] LeCun Y, Bengio Y, Hinton G . Deep Learning[J]. Nature, 2015,521(7553):436-444.
[73] Greff K, Srivastava R K, Koutník J , et al. LSTM: A Search Space Odyssey[J]. IEEE Transactions on Neural Networks and Learning Systems, 2017,28(10):2222-2232.
[74] Schmidhuber J . Deep Learning in Neural Networks: An Overview[J]. Neural Networks, 2015,61:85-117.
[75] Olah C . Understanding LSTM Networks[OL].(2015-08-27). [ 2017-10-15]. http://colah.github.io/posts/2015-08-Understanding-LSTMs/. .
[76] Wang Y, Huang M, Zhao L. Attention-Based LSTM for Aspect-Level Sentiment Classification [C]// Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, Austin, Texas, USA. 2016: 606-615.
[77] Page L, Brin S, Motwani R , et al. The PageRank Citation Ranking: Bringing Order to The Web[OL].(2001-10-30). [ 2017-10-15]. http://ilpubs.stanford.edu:8090/422/.
[78] Verrecchia R E . Information Acquisition in a Noisy Rational Expectations Economy[J]. Econometrica: Journal of the Econometric Society, 1982,50(6):1415-1430.
[79] Merton R C . A Simple Model of Capital Market Equilibrium with Incomplete Information[J]. The Journal of Finance, 1987,42(3):483-510.
[80] DellaVigna S, Pollet J M . Investor Inattention and Friday Earnings Announcements[J]. The Journal of Finance, 2009,64(2):709-749.
[81] Barber B M, Odean T . All That Glitters: The Effect of Attention and News on the Buying Behavior of Individual and Institutional Investors[J]. The Review of Financial Studies, 2007,21(2):785-818.
[82] Tetlock P C . All the News That’s Fit to Reprint: Do Investors React to Stale Information?[J]. The Review of Financial Studies, 2011,24(5):1481-1512.
[1] 卢伟聪,徐健. 基于三分网络的网络用户评论情感分析 *[J]. 数据分析与知识发现, 2019, 3(8): 10-20.
[2] 尤众喜,华薇娜,潘雪莲. 中文分词器对图书评论和情感词典匹配程度的影响 *[J]. 数据分析与知识发现, 2019, 3(7): 23-33.
[3] 蒋翠清,郭轶博,刘尧. 基于中文社交媒体文本的领域情感词典构建方法研究*[J]. 数据分析与知识发现, 2019, 3(2): 98-107.
[4] 余本功,张培行,许庆堂. 基于F-BiGRU情感分析的产品选择方法*[J]. 数据分析与知识发现, 2018, 2(9): 22-30.
[5] 曾子明,杨倩雯. 基于LDA和AdaBoost多特征组合的微博情感分析*[J]. 数据分析与知识发现, 2018, 2(8): 51-59.
[6] 何有世,何述芳. 基于领域本体的产品网络口碑信息多层次细粒度情感挖掘*[J]. 数据分析与知识发现, 2018, 2(8): 60-68.
[7] 张宁,尹乐民,何立峰. 网络股评“发布者-关注者”BSI与股票市场关联性研究*[J]. 数据分析与知识发现, 2018, 2(6): 1-12.
[8] 王秀芳,盛姝,路燕. 一种基于话题聚类及情感强度的微博舆情分析模型*[J]. 数据分析与知识发现, 2018, 2(6): 37-47.
[9] 杨斯楠,徐健,叶萍萍. 网络评论情感可视化技术方法及工具研究*[J]. 数据分析与知识发现, 2018, 2(5): 77-87.
[10] 王婷婷,王凯平,戚桂杰. 基于情感分析的开放式创新平台创意采纳研究: 以Salesforce为例*[J]. 数据分析与知识发现, 2018, 2(4): 38-47.
[11] 赵杨,李齐齐,陈雨涵,曹文航. 基于在线评论情感分析的海淘APP用户满意度研究*[J]. 数据分析与知识发现, 2018, 2(11): 19-27.
[12] 何跃,朱灿. 基于微博的意见领袖网情感特征分析*——以“非法疫苗”事件为例[J]. 数据分析与知识发现, 2017, 1(9): 65-73.
[13] 张红丽,刘济郢,杨斯楠,徐健. 基于网络用户评论的评分预测模型研究*[J]. 数据分析与知识发现, 2017, 1(8): 48-58.
[14] 高歌,罗珺玫,王宇. 基于HNC理论的文本情感倾向性分析*[J]. 数据分析与知识发现, 2017, 1(8): 85-91.
[15] 首欢容,邓淑卿,徐健. 基于情感分析的网络谣言识别方法*[J]. 数据分析与知识发现, 2017, 1(7): 44-51.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
版权所有 © 2015 《数据分析与知识发现》编辑部
地址:北京市海淀区中关村北四环西路33号 邮编:100190
电话/传真:(010)82626611-6626,82624938
E-mail:jishu@mail.las.ac.cn