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

REN Xiaoping; CHEN Zhiping

doi:10.7652/xjtuxb202502015

您当前的位置：

首页 >

文章列表页 >

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

更新时间：2026-01-06

- Quantitative Model for Stock Indexes Based on Gated Recurrent Unit and Deep Evolution Strategy
- Journal of Xi’an Jiaotong University Vol. 59, Issue 2, Pages: 146-155(2025)
- 作者机构：
  
  1.西安交通大学数学与统计学院， 710049，西安
  2.东北石油大学数学与统计学院， 163318，黑龙江大庆
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb202502015
  CLC： TP181
- Received：23 May 2024，
  
  Online First：24 October 2024，
  
  Published：10 February 2025
- 稿件说明：
移动端阅览
REN Xiaoping, CHEN Zhiping. Quantitative Model for Stock Indexes Based on Gated Recurrent Unit and Deep Evolution Strategy[J]. Journal of Xi’an Jiaotong University, 2025, 59(2): 146-155.
DOI：

REN Xiaoping, CHEN Zhiping. Quantitative Model for Stock Indexes Based on Gated Recurrent Unit and Deep Evolution Strategy[J]. Journal of Xi’an Jiaotong University, 2025, 59(2): 146-155. DOI： 10.7652/xjtuxb202502015.

摘要

为了提高股票价指数预测的准确性、增强统计建模性能优化与股票指数特征相依的交易策略效果，提出一种将指数预测与量化交易策略有效结合的门控循环单元深度进化量化模型（GRU-DES）。首先，建立循环神经网络（RNN）、长短时记忆神经网络（LSTM）和门控循环单元网络（GRU）预测模型，分别对上海证券交易所（上证）超大盘股票指数、上证中盘股票指数和上证小盘股票指数进行预测；接着采用所提出的深度进化量化模型（DES）对三大股票指数的预测值与真实值进行回测研究，通过比较预测结果与真实结果在同一策略下的各项回测指标和交易细节等特性确定最优网络结构和策略参数，进而优化深度进化策略；最后根据优化后的策略提出了GRU-DES模型，并再次对三大股票指数进行样本外数据回测来验证模型有效性。实证回测结果表明：所提出的GRU-DES模型在各量化回测指标上较LSTM-DES模型与RNN-DES模型的预测精度均高出14%以上，有效解决了统计预测指标的随机性和过拟合的问题；根据2016年至2024年7年间数据回测，所提出的GRU-DES模型比强化学习模型在各回测指标中均展现了稳定性和有效性。

Abstract

To improve the prediction accuracy of stock price indexes and enhance the performance of statistical modeling and the design of quantitative trading strategy based on the characteristics of indexes

a quantitative model was proposed

which was based on gated recurrent unit and deep evolution strategy (GRU-DES) and effectively integrated index prediction with quantitative trading strategies. Firstly

RNN

LSTM

and GRU neural network predictive models were established to forecast the SSE mega cap

SSE mid cap and SSE small cap indexes

respectively. Next

the proposed DES model was employed to backtest the predicted values and true values of the three indexes. By comprehensively comparing the backtesting metrics and trading details of the predicted results with the true results under the same strategy

after determining the network structure and strategy parameters

the DES was optimized. Finally

the GRU-DES model was developed on basis of the optimized strategy

and the model effectiveness was verified through out-of-sample backtesting of these indexes again. The results show that the proposed GRU-DES model was 14% higher than the LSTM-DES model and RNN-DES model in backtesting metrics

effectively avoiding the randomness and overfitting problems of statistical prediction indexes. According to the backtesting results over 7 years from 2016 to 2024

the proposed GRU-DES model sufficiently demonstrates the stability and effectiveness in all backtesting metrics compared to reinforcement learning model.

关键词

Keywords

references

AYALA J , GARCÍA-TORRES M , NOGUERA J L V , et al . Technical analysis strategy optimization using a machine learning approach in stock market indices [J ] . Knowledge-Based Systems , 2021 , 225 : 107119 .

WANG Jia , WANG Xinyi , WANG Xu . International oil shocks and the volatility forecasting of Chinese stock market based on machine learning combination models [J ] . The North American Journal of Economics and Finance , 2024 , 70 : 102065 .

PENG Zongyu , GUO Peichang . A data organization method for LSTM and transformer when predicting Chinese banking stock prices [J ] . Discrete Dynamics in Nature and Society , 2022 , 2022 ( 1 ): 7119678 .

LIN Zili , TIAN Fangyuan , ZHANG Weiqian . Evaluation and analysis of an LSTM and GRU based stock investment strategy [C ] // Proceedings of the 2022 International Conference on Economics, Smart Finance and Contemporary Trade (ESFCT 2022) . Dordrecht, The Netherlands : Atlantis Press , 2022 : 1615 - 1626 .

SANG Chenjie , DI PIERRO M . Improving trading technical analysis with TensorFlow long short-term memory (LSTM) neural network [J ] . The Journal of Finance and Data Science , 2019 , 5 ( 1 ): 1 - 11 .

KANWAL A , LAU M F , NG S P H , et al . BiCuDNNLSTM-1dCNN: a hybrid deep learning-based predictive model for stock price prediction [J ] . Expert Systems with Applications , 2022 , 202 : 117123 .

LV Pin , SHU Yating , XU Jia , et al . Modal decomposition-based hybrid model for stock index prediction [J ] . Expert Systems with Applications , 2022 , 202 : 117252 .

于孝建 , 刘国鹏 , 刘建林 , 等 . 基于LSTM网络和文本情感分析的股票指数预测 [J ] . 中国管理科学 , 2024 , 32 ( 8 ): 25 - 35 .

YU Xiaojian , LIU Guopeng , LIU Jianlin , et al . Stock index prediction based on LSTM network and text sentiment analysis [J ] . Chinese Journal of Management Science , 2024 , 32 ( 8 ): 25 - 35 .

LIU Yijiao , LIU Xinghua , ZHANG Yuxin , et al . CEGH: a hybrid model using CEEMD, Entropy, GRU, and history attention for intraday stock market forecasting [J ] . Entropy , 2023 , 25 ( 1 ): 71 .

周章元 , 何小灵 . 基于优化LSTM模型的股价预测方法 [J ] . 统计与决策 , 2023 , 39 ( 6 ): 143 - 148 .

ZHOU Zhangyuan , HE Xiaoling . Stock price prediction method based on optimized LSTM model [J ] . Statistics & Decision , 2023 , 39 ( 6 ): 143 - 148 .

BIŃKOWSKI M , MARTI G , DONNAT P . Autoregressive convolutional neural networks for asynchronous time series [C ] // Proceedings of the 35th International Conference on Machine Learning . Chia Laguna Resort, Sardinia, Italy : PMLR , 2018 : 580 - 589 .

LUO Shuang , NI Zhiwei , ZHU Xuhui , et al . A novel methanol futures price prediction method based on multicycle CNN-GRU and attention mechanism [J ] . Arabian Journal for Science and Engineering , 2023 , 48 ( 2 ): 1487 - 1501 .

NOURBAKHSH Z , HABIBI N . Combining LSTM and CNN methods and fundamental analysis for stock price trend prediction [J ] . Multimedia Tools and Applications , 2023 , 82 ( 12 ): 17769 - 17799 .

LI Chengyu , QIAN Guoqi . Stock price prediction using a frequency decomposition based GRU transformer neural network [J ] . Applied Sciences , 2023 , 13 ( 1 ): 222 .

KHURSHID B , MAQSOOD S . A hybrid evolution strategies-simulated annealing algorithm for job shop scheduling problems [J ] . Engineering Applications of Artificial Intelligence , 2024 , 133 , Part A : 108016 .

ABOYEJI E T , AJANI O S , MALLIPEDDI R . Covariance matrix adaptation evolution strategy based on ensemble of mutations for parking navigation and maneuver of autonomous vehicles [J ] . Expert Systems with Applications , 2024 , 249 , Part A : 123565 .

NEMATZADEH H , GARCÍA-NIETOJ , ALDANA-MONTES J F , et al . Pattern recognition frequency-based feature selection with multi-objective discrete evolution strategy for high-dimensional medical datasets [J ] . Expert Systems with Applications , 2024 , 249 , Part A : 123521 .

LO A W , SINGH M . Deep-learning models for forecasting financial risk premia and their interpretations [J ] . Quantitative Finance , 2023 , 23 ( 6 ): 917 - 929 .

刘伟龙 , 张永 , 杨兴雨 . 基于LSTM预测信息的在线融资融券组合交易策略 [J ] . 系统工程理论与实践 , 2024 , 44 ( 8 ): 2493 - 2508 .

LIU Weilong , ZHANG Yong , YANG Xingyu . Online margin trading strategy based on LSTM prediction information [J ] . Systems Engineering-Theory & Practice , 2024 , 44 ( 8 ): 2493 - 2508 .

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 .

柴昱白 , 陈伟 , 赵舒欣 , 等 . 采用机器学习与二维伽马函数的股票指数量化交易策略 [J ] . 西安交通大学学报 , 2023 , 57 ( 5 ): 204 - 212 .

CHAI Yubai , CHEN Wei , ZHAO Shuxin , et al . Quantitative trading strategy for stock index based on machining learning and 2D gamma function [J ] . Journal of Xi'an Jiaotong University , 2023 , 57 ( 5 ): 204 - 212 .

贺毅岳 , 刘磊 , 高妮 . 基于M-LSTM的股票指数日内交易量分布预测研究 [J ] . 运筹与管理 , 2022 , 31 ( 10 ): 196 - 203 .

HE Yiyue , LIU Lei , GAO Ni . Research on forecasting intraday trading volume of stock index based on M-LSTM [J ] . Operations Research and Management Science , 2022 , 31 ( 10 ): 196 - 203 .

GUPTA U , BHATTACHARJEE V , BISHNU P S . StockNet-GRU based stock index prediction [J ] . Expert Systems with Applications , 2022 , 207 : 117986 .

蔡超 , 何馨怡 , 李丽 . 大规模数据的分布式神经网络回归模型研究 [J ] . 统计与决策 , 2023 , 39 ( 17 ): 34 - 39 .

CAI Chao , HE Xinyi , LI Li . Research on distributed neural network regression model for large-scale data [J ] . Statistics & Decision , 2023 , 39 ( 17 ): 34 - 39 .

KU C S , XIONG Jiale , CHEN Y L , et al . Improving stock market predictions: an equity forecasting scanner using long short-term memory method with dynamic indicators for Malaysia stock market [J ] . Mathematics , 2023 , 11 ( 11 ): 2470 .

NING B , LIN F H T , JAIMUNGAL S . Double deep Q-learning for optimal execution [J ] . Applied Mathematical Finance , 2021 , 28 ( 4 ): 361 - 380 .

XU Hongfeng , CHAI Lei , LUO Zhiming , et al . Stock movement prediction via gated recurrent unit network based on reinforcement learning with incorporated attention mechanisms [J ] . Neurocomputing , 2022 , 467 : 214 - 228 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Equipment Remaining Useful Life Prediction Method Using Multi-Scale Mamba Architecture

Application of Mayfly Algorithm Optimized Dual-Channel Network in Gearbox Fault Diagnosis

A Solid-State Drive Fault Detection Method with Dimensionality Reduction

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

Bi-Level Multi-Obj ective Optimization of a Wind-Solar Coupled Hydrogen Production System Based on the Synergy of Hybrid Electrolyzer Arrays

Related Author

KONG Xiangyu

ZHENG Ziyang

FAN Hongdong

HU Changhua

WANG Zhaoqiang

TIAN Congbao

WEI Chenqian

YU Xiaohan

Related Institution

Laboratory of Intelligent Control, Rocket Force University of Engineering

State Key Laboratory of Fluid Power Components and Mechatronic Systems, Zhejiang University

College of Mechanical Engineering, Liaoning Technical University

(Faculty of Electronic and Information Engineering, Xi'an Jiaotong University,,)

⁰