Please wait a minute...
浙江大学学报(理学版)  2019, Vol. 46 Issue (5): 550-555    DOI: 10.3785/j.issn.1008-9497.2019.05.008
数学与计算机科学     
基于神经网络的股票收益率预测研究
潘水洋, 刘俊玮, 王一鸣
北京大学 经济学院,北京 100871
Forecasting stock returns with artificial neural networks.
PAN Shuiyang, LIU Junwei, WANG Yiming
School of Economics, Peking University, Beijing 100871,China
 全文: PDF(439 KB)   HTML  
摘要: 在金融领域的资产定价模型修正过程中,股市的非线性现象往往被选择性忽视,未纳入模型框架,现有模型亦无法刻画因子之间的非线性定价结构。为解决上述问题,引入了机器学习领域中的神经网络模型,以捕获市场组合收益率、市值、账面市值比三因子间的非线性定价结构,并对股票收益率进行预测。将该模型与经典Fama-French三因子模型在样本外拟合优度、多空策略业绩表现上做了对比,结果表明:神经网络模型能精准捕获市场组合收益率、市值、账面市值比3个因子之间的非线性关系,且在样本外拟合优度、多空策略业绩表现上均要优于传统三因子线性定价模型。
关键词: 机器学习非线性资产定价神经网络    
Abstract: In the process of correcting the asset pricing model, the stock market exhibits nonlinear phenomena.Nevertheless this widespread financial disparity is ignored selectively and not included in the model framework. The challenge is that the existing model cannot describe the nonlinear pricing structure among factors. In this paper, a neural network model in the field of machine learning is presented to capture the nonlinear pricing structure between the three factors of firm size, book-to-market ratio and return on the market portfolio. Based on the Chinese A-share market data, this model is fully compared with the classic Fama-French three-factor model. The empirical results show that the neural network model is better than the Fama-French three-factor linear pricing model in terms of goodness of fit outside the sample and performance of long-short strategy.
Key words: machine learning    nonlinear asset pricing    neural network
收稿日期: 2018-12-26 出版日期: 2019-09-25
CLC:  O212  
基金资助: 国家社会科学基金青年项目(18CJY057).
作者简介: 潘水洋(1986―),ORCID:http:// orcid.org/0000-0003-3512-6503 , 男,博士后,主要从事实证资产定价与金融大数据分析研究. E-mail:panshuiyang@126.com.
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  
潘水洋
刘俊玮
王一鸣

引用本文:

潘水洋, 刘俊玮, 王一鸣. 基于神经网络的股票收益率预测研究[J]. 浙江大学学报(理学版), 2019, 46(5): 550-555.

PAN Shuiyang, LIU Junwei, WANG Yiming. Forecasting stock returns with artificial neural networks.. Journal of ZheJIang University(Science Edition), 2019, 46(5): 550-555.

链接本文:

https://www.zjujournals.com/sci/CN/10.3785/j.issn.1008-9497.2019.05.008        https://www.zjujournals.com/sci/CN/Y2019/V46/I5/550

1 FAMAE F, FRENCHK R. The cross-section of expected stock returns[J]. The Journal of Finance, 1992,47(2):427-465.
2 FAMAE F, FRENCHK R. A five-factor asset pricing model[J]. Journal of Financial Economics, 2015,116(1):1-22.
3 ZHANGL. The investment CAPM[J]. European Financial Management, 2017,23(4):545-603.DOI:10.1111/eufm.12129
4 ADRIANT, FRANZONIF. Learning about Beta: Time-varying factor loadings, expected returns, and the conditional CAPM[J]. Journal of Empirical Finance, 2009,16(4):537-556.DOI:10.2139/ssrn.391562
5 HARVEYC R, LIUY, ZHUH Q.…and the cross-section of expected returns[J]. The Review of Financial Studies, 2016,29(1):5-68.DOI:10.2139/ssrn.2249314
6 LEWELLENJ, NAGELS. The conditional CAPM does not explain asset-pricing anomalies[J]. Journal of Financial Economics,2006,82(2):289-314.DOI:10.2139/ssrn.441341
7 DAVIDSONJ, TERÄSVIRTAT. Long memory and nonlinear time series[J]. Journal of Econometrics,2002,110(2):105-112.
8 HAOQ M. R/S nonlinear analysis on the long memory of Chinese stock markets return[J]. Journal of Industrial Engineering Management, 2007,21(2):115-117.
9 LIH Q, WANGS Y, MAC Q. What’s the nature of volatility in stock prices? -Based on the nonlinear dynamical analysis principle[J]. Chinese Journal of Management Science, 2008,16(5):1-8.
10 PANASE. Estimating fractal dimension using stable distributions and exploring long memory through ARFIMA models in Athens stock exchange[J]. Applied Financial Economics, 2001,11(4):395-402.DOI:10.1080/096031001300313956
11 ATHEYS. Beyond prediction: Using big data for policy problems[J]. Science,2017, 355(6324):483-485.DOI:10.1126/science.aal4321
12 VARIANR. Big data: New tricks for econometrics[J]. Journal of Economic Perspectives, 2014,28(2): 3-28.DOI:10.1257/jep.28.2.3
13 VARIANH R. Causal inference in economics and marketing[J]. Proceedings of the National Academy of Sciencess,2016,113(27):7310-7315. DOI:10.1073/pnas.1510479113
14 HOUK W, XUEC, ZHANGL. Digesting anomalies: An investment approach[J]. The Review of Financial Studies, 2015,28(3): 650-705. DOI:10.2139/ssrn.2152674
15 ZHAOS M, YANH L, ZHANGK. Does Fama-French five factor model outperform three factor model? Evidence from China's A-share market[J]. Nankai Economic Studies, 2016,15(2):41-59.
[1] 郑素佩,靳放,封建湖,林云云. 双曲型方程激波捕捉的物理信息神经网络(PINN)算法[J]. 浙江大学学报(理学版), 2023, 50(1): 56-62.
[2] 吴郁文,林杰. 融合遥感与社会感知数据的城市土地利用分类方法[J]. 浙江大学学报(理学版), 2023, 50(1): 83-95.
[3] 常晓洁,张华. 一种基于V-TGRU模型的资源调度算法[J]. 浙江大学学报(理学版), 2022, 49(4): 467-473.
[4] 郑素佩,闫佳,宋学力,陈荧. 求解大规模矛盾方程组的最小二乘支持向量机算法[J]. 浙江大学学报(理学版), 2022, 49(4): 435-442.
[5] 刘华玲,张国祥,马俊. 图嵌入算法研究进展[J]. 浙江大学学报(理学版), 2022, 49(4): 443-456.
[6] 郏丽丽,孙婷婷. 紫色球杆菌视紫红质光谱特性的机器学习研究[J]. 浙江大学学报(理学版), 2022, 49(3): 280-286.
[7] 王昱文,杜震洪,戴震,刘仁义,张丰. 基于复合神经网络的多元水质指标预测模型[J]. 浙江大学学报(理学版), 2022, 49(3): 354-362.
[8] 郭毅博, 牛猛, 王海迪, 陈艳华, 薛均晓, 袁玥, 侯立硕, 徐明亮, 潘俊. 基于生成对抗网络的飞机燃油数据缺失值填充方法[J]. 浙江大学学报(理学版), 2021, 48(4): 402-409.
[9] 王小超, 张雷, 余元强, 胡坤, 胡建平. 基于快速神经网络架构搜索的鲁棒图像水印网络算法[J]. 浙江大学学报(理学版), 2021, 48(3): 261-269.
[10] 王协, 章孝灿, 苏程. 基于多尺度学习与深度卷积神经网络的遥感图像土地利用分类[J]. 浙江大学学报(理学版), 2020, 47(6): 715-723.
[11] 曾金迪, 张丰, 吴森森, 杜震洪, 刘仁义. 基于空间自回归神经网络模型的空间插值研究[J]. 浙江大学学报(理学版), 2020, 47(5): 572-581.
[12] 李君轶, 任涛, 陆路正. 游客情感计算的文本大数据挖掘方法比较研究[J]. 浙江大学学报(理学版), 2020, 47(4): 507-520.
[13] 周笑天, 张丰, 杜震洪, 刘仁义. 基于神经网络集合预报的台风路径预报优化[J]. 浙江大学学报(理学版), 2020, 47(2): 196-202.
[14] 卢家品, 罗月童, 黄兆嵩, 张延孔, 陈为. 基于排名学习和多源信息的地图匹配方法[J]. 浙江大学学报(理学版), 2020, 47(1): 27-35.
[15] 刘尧, 王颖志, 王立君, 张丰, 杜震洪, 刘仁义. 交通事故的时空热点分析[J]. 浙江大学学报(理学版), 2020, 47(1): 52-59.