Quantitative Trading Strategy for Stock Index Based on Machining Learning and 2D Gamma Function

CHAI Yubai; CHEN Wei; ZHAO Shuxin; MAO Xinyue

doi:10.7652/xjtuxb202305020

您当前的位置：

首页 >

文章列表页 >

Quantitative Trading Strategy for Stock Index Based on Machining Learning and 2D Gamma Function

更新时间：2025-07-09

- Quantitative Trading Strategy for Stock Index Based on Machining Learning and 2D Gamma Function
- Vol. 57, Issue 5, Pages: 204-212(2023)
- 作者机构：
  
  西安交通大学电子与信息学部,西安,710049
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb202305020
  CLC： F832.51;TP181
- Online First：10 May 2023，
  
  Published：2023
- 稿件说明：
移动端阅览
CHAI Yubai, CHEN Wei, ZHAO Shuxin, et al. Quantitative Trading Strategy for Stock Index Based on Machining Learning and 2D Gamma Function[J]. 2023, 57(5): 204-212.
DOI：

CHAI Yubai, CHEN Wei, ZHAO Shuxin, et al. Quantitative Trading Strategy for Stock Index Based on Machining Learning and 2D Gamma Function[J]. 2023, 57(5): 204-212. DOI： 10.7652/xjtuxb202305020.

摘要

为了通过短线交易获取对标股票指数的超额收益

提出一种基于机器学习的股票指数增强型量化交易策略

并实现程序化自动交易。首先生成股票指数的初始特征集

通过最大互信息系数法筛选得21维特征; 然后设计双阈值涨跌不平衡标签与3种预测分类方式(T、B-B、B-T)

组合构建不同机器学习模型

并对比择优得到多空交易方向的最优机器学习模型分别为LSTM-B-T模型与RF-B-T模型; 接着设计基于日内涨跌幅的二维伽马函数; 最后使用经二维伽马函数计算得到的类概率判别阈值与最优机器学习模型预测的类概率进行涨跌类别判定

得到交易信号。将该策略应用于中证500指数的股指ETF进行回测与模拟盘交易验证

实验结果表明:相较于随机建仓策略

采用该策略使交易评价指标得到整体性提升; 在回测与为期3个月的模拟盘交易验证中使用该策略均能获得对标指数的理想超额收益

分别为11.24%、11.08%。

Abstract

This paper proposes a stock index enhanced quantitative trading strategy based on machine learning and realizes programmed automatic trading to obtain the excess return of the benchmark stock index through short-term trading. First

the initial feature set of the stock index is generated

and the 21-dimensional features are screened by the maximum information coefficient method. Then a double-threshold unbalanced label and three prediction classification methods(T

B-B

B-T)are designed. In addition

different machine learning models are built and compared to obtain the optimal machine learning models for the long-short trading direction: LSTM-B-T model and RF-B-T model respectively. After that

a two-dimensional gamma function based on intraday rise and fall is designed. Finally

ups and downs are determined and trading signals obtained using the class probability discrimination threshold calculated by the two-dimensional gamma function and the class probability predicted by the optimal machine learning models. This strategy is applied to the backtesting and simulation trading verification of the CSI 500 ETF. The experimental results show that

compared with the random position opening strategy

the trading strategy proposed in this paper makes the transaction evaluation index improve on the whole. This strategy can obtain the ideal excess return of the benchmark index in both backtesting and three-month simulation trading verification

which are 11.24% and 11.08% respectively.

关键词

Keywords

references

陈伟忠, 金以萍, 汪应洛. 证券资产动态定价模型与机制研究 [J]. 西安交通大学学报, 1997, 31(10): 107-113.

CHEN Weizhong, JIN Yiping, WANG Yingluo. A capital asset pricing model for emerging market with changeable opportunity set [J]. Journal of Xi'an Jiaotong University, 1997, 31(10): 107-113.

NEELY C J, RAPACH D E, TU Jun, et al. Forecasting the equity risk premium: the role of technical indicators [J]. Management Science, 2014, 60(7): 1772-1791.

CHEN Yiwen, CHOU R K, LIN Chubin. Investor sentiment, SEO market timing, and stock price performance [J]. Journal of Empirical Finance, 2019, 51: 28-43.

LOHRMANN C, LUUKKA P. Classification of intraday SP500 returns with a random Forest [J]. International Journal of Forecasting, 2019, 35(1): 390-407.

王淑燕, 曹正凤, 陈铭芷. 随机森林在量化选股中的应用研究 [J]. 运筹与管理, 2016, 25(3): 163-168, 177.

WANG Shuyan, CAO Zhengfeng, CHEN Mingzhi. Research on application of random forests in the quantitative stock selection model [J]. Operations Research and Management Science, 2016, 25(3): 163-168, 177.

李斌, 林彦, 唐闻轩. ML-TEA: 一套基于机器学习和技术分析的量化投资算法 [J]. 系统工程理论与实践, 2017, 37(5): 1089-1100.

LI Bin, LIN Yan, TANG Wenxuan. ML-TEA: a set of quantitative investment algorithms based on machine learning and technical analysis [J]. Systems Engineering-Theory Practice, 2017, 37(5): 1089-1100.

张茂军, 饶华城, 南江霞, 等. 基于决策树的量化交易择时策略 [J]. 系统工程, 2022, 40(2): 118-130.

ZHANG Maojun, RAO Huacheng, NAN Jiangxia, et al. Timing selection of quantitative trading based on decision tree [J]. Systems Engineering, 2022, 40(2): 118-130.

高妮, 贺毅岳, 马新成. 基于低频分量EEMD-SVR预测建模的指数择时策略 [J]. 统计与决策, 2022, 38(2): 140-145.

GAO Ni, HE Yiyue, MA Xincheng. Exponential timing strategy based on EEMD-SVR predictive modeling of low frequency component [J]. Statistics Decision, 2022, 38(2): 140-145.

NEMES M D, BUTOI A. Data mining on Romanian stock market using neural networks for price prediction [J]. Informatica Economica Journal, 2013, 17(3): 125-136.

FISCHER T, KRAUSS C. Deep learning with long short-term memory networks for financial market predictions [J]. European Journal of Operational Research, 2018, 270(2): 654-669.

杨青, 王晨蔚. 基于深度学习LSTM神经网络的全球股票指数预测研究 [J]. 统计研究, 2019, 36(3): 65-77.

YANG Qing, WANG Chenwei. A study on forecast of global stock indices based on deep LSTM neural network [J]. Statistical Research, 2019, 36(3): 65-77.

贺毅岳, 高妮, 韩进博, 等. 基于长短记忆网络的指数量化择时研究 [J]. 统计与决策, 2020, 36(23): 128-133.

HE Yiyue, GAO Ni, HAN Jinbo, et al. Research on the exponential quantitative timing based on long short term memory [J]. Statistics Decision, 2020, 36(23): 128-133.

陈玲, 琚亚平, 张楚华. 多级轴流压气机末级静子叶片串列改型方法及流动机理研究 [J]. 西安交通大学学报, 2022, 56(10): 170-179.

CHEN Ling, JU Yaping, ZHANG Chuhua. Tandem blading and flow mechanism of last stator of multi-stage axial flow compressor [J]. Journal of Xi'an Jiaotong University,2022, 56(10): 170-179.

张西宁, 张雯雯, 周融通, 等. 采用单类随机森林的异常检测方法及应用 [J]. 西安交通大学学报, 2020, 54(2): 1-8, 157.

ZHANG Xining, ZHANG Wenwen, ZHOU Rongtong, et al. Anomaly detection method based on one-class random forest with applications [J]. Journal of Xi'an Jiaotong University, 2020, 54(2): 1-8, 157.

陈保家, 陈学力, 沈保明, 等. CNN-LSTM深度神经网络在滚动轴承故障诊断中的应用 [J]. 西安交通大学学报, 2021, 55(6): 28-36.

CHEN Baojia, CHEN Xueli, SHEN Baoming, et al. An application of convolution neural network and long short-term memory in rolling bearing fault diagnosis [J]. Journal of Xi'an Jiaotong University, 2021, 55(6): 28-36.

任君, 王建华, 王传美, 等. 基于正则化LSTM模型的股票指数预测 [J]. 计算机应用与软件, 2018, 35(4): 44-48, 108.

REN Jun, WANG Jianhua, WANG Chuanmei, et al. Stock index forecast based on regularized LSTM model [J]. Computer Applications and Software, 2018, 35(4): 44-48, 108.

PASCANU R, MIKOLOV T, BENGIO Y. On the difficulty of training recurrent neural networks [C]//Proceedings of the 30th International Conference on Machine Learning. Pittsburgh, PA, USA: IMLS, 2013: 2347-2355.

陈佳, 刘冬雪, 武大硕. 基于特征选取与LSTM模型的股指预测方法研究 [J]. 计算机工程与应用, 2019, 55(6): 108-112.

CHEN Jia, LIU Dongxue, WU Dashuo. Stock index forecasting method based on feature selection and LSTM model [J]. Computer Engineering and Applications, 2019, 55(6): 108-112.

RESHEF D N, RESHEF Y A, FINUCANE H K, et al. Detecting novel associations in large data sets [J]. Science, 2011, 334(6062): 1518-1524.

张园园. 基于资金流及股指短期指数变化的程序化交易软件设计与实现 [D]. 西安: 西安交通大学, 2021.

陆珩瑱, 姚恒宇. 涨跌幅限制变动对市场波动性及流动性影响的仿真研究 [J]. 系统工程理论与实践, 2022, 42(4): 897-905.

LU Hengzhen, YAO Hengyu. A simulation study on the effect of price limits change on market volatility and liquidity [J]. Systems Engineering-Theory Practice, 2022, 42(4): 897-905.

李梦雨. 证券价格涨跌幅限制制度的存在原因:基于反市场操纵的视角 [J]. 西安交通大学学报(社会科学版), 2015, 35(1): 34-39.

LI Mengyu. Reasons for existence of system of securities price limits: from a visual angle against market manipulation [J]. Journal of Xi'an Jiaotong University(Social Sciences), 2015, 35(1): 34-39.

刘志成, 王殿伟, 刘颖, 等. 基于二维伽马函数的光照不均匀图像自适应校正算法 [J]. 北京理工大学学报, 2016, 36(2): 191-196, 214.

LIU Zhicheng, WANG Dianwei, LIU Ying, et al. Adaptive adjustment algorithm for non-uniform illumination images based on 2D gamma function [J]. Transactions of Beijing Institute of Technology, 2016, 36(2): 191-196, 214.

MANURUNG A H, BUDIHARTO W, PRABOWO H. Algorithm and modeling of stock prices forecasting based on long short-term memory(LSTM)[J]. ICIC Express Letters, 2018, 12(12): 1277-1283.

KRAUSS C, DO X A, HUCK N. Deep neural networks, gradient-boosted trees, random forests: statistical arbitrage on the SP 500 [J]. European Journal of Operational Research, 2017, 259(2): 689-702.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Advances in the Application of Machine Learning to Boiling Heat Transfer

Machine Learning-Assisted Optimization of Generalized k-ω Turbulence Model for Transonic Compressor Rotors

Research on Adsorbents for CO2 Capture：From Traditional Design to Machine Learning

Quantitative Model for Stock Indexes Based on Gated Recurrent Unit and Deep Evolution Strategy

Knowledge Graph Representation Learning Driven by Semantic Discrimination and Subgraph Integration

Related Author

PAVLENKO Alexander Nikolaevich

WANG Qiuwang

SURTAEV Anton Sergeevich

ZHOU Chenjie

BI Xiaolong

TANG Weifeng

CHU Wenxiao

LI Zhen

Related Institution

Kutateladze Institute of Thermophysics，Siberian Branch of the Russian Academy of Sciences

MOE Key Laboratory of Thermo-Fluid Science and Engineering，Xi’an Jiaotong University

School of Energy and Power Engineering, Xi'an Jiaotong University

中船（邯郸)派瑞特种气体股份有限公司，056000，河北邯郸Peric Special Gases Co., Ltd., Handan, Hebei 056000, China

西安交通大学能源与动力工程学院，710049，西安School of Energy and Power Engineering, Xi'an Jiaotong University, Xi'an 710049, China

⁰