人工智能在金融领域的数据预测方法综述

doi:10.11896/jsjkx.250700166

摘要/Abstract

摘要： 面对金融市场的高度复杂性与数据的高噪声、非线性特性,以机器学习和深度学习为代表的人工智能技术已成为金融数据预测领域的核心驱动力。文中系统性地梳理与总结了近3年人工智能在该领域的最新研究进展与核心方法论。在传统机器学习层面,研究趋势已从单一学习器的应用,发展至以堆叠(Stacking)为代表的模型融合策略,以及结合优化算法进行自动化特征选择与超参数调优的混合范式。在深度学习层面,其应用展现出一条清晰的演进路径,即从作为基石的循环神经网络(RNN)及其与数据分解、注意力机制的融合,到能够捕捉多维度信息的CNN-RNN混合架构;并进一步聚焦于更前沿的图神经网络(GNN)、Transformer架构以及深度强化学习(DRL),系统阐述了它们如何分别在网络关联建模、长序列依赖捕捉以及实现“预测到决策”的范式转变中展现出独特优势。此外,还归纳了关键的特征工程与数据处理技术,并创新性地提出一个旨在反映金融产品内在拓扑结构的“结构化神经网络建模”新范式,以增强模型的可解释性。最后,在对现有成果进行总结的基础上,指出了当前研究在数据质量、模型可解释性与鲁棒性等方面面临的核心挑战,并对深度多模态融合、因果推断、金融大语言模型(LLM)以及可信赖AI(XAI)等未来关键研究方向进行了展望,以期为相关领域的学术研究与业界实践提供全面且具有前瞻性的参考。

关键词: 金融预测, 时间序列, 投资组合管理, 人工智能, 机器学习, 深度学习

Abstract: Given the high complexity of financial markets and the inherently noisy,non-linear nature of their data,artificial intelligence(AI),particularly machine learning and deep learning,has emerged as a core driving force in financial data prediction.This paper systematically summarizes the latest research progress and core methodologies in this domain over the past three years.In traditional machine learning,the trend has shifted from applying single learners toward sophisticated model fusion strategies,such as Stacking,and hybrid paradigms that integrate optimization algorithms for automated feature selection and hyperparameter tu-ning.Deep learning applications demonstrate a clear evolutionary trajectory,starting with foundational Recurrent Neural Networks(RNNs) enhanced by data decomposition and attention mechanisms,and progressing to hybrid architectures like CNN-RNN for capturing multi-dimensional features.This paper further details the adoption of cutting-edge models:Graph Neural Networks(GNNs) for modeling entity relationships,Transformers for capturing long-range dependencies,and Deep Reinforcement Lear-ning(DRL) for shifting the paradigm from prediction to autonomous decision-making.Furthermore,the review outlines key feature engineering techniques and introduces an innovative “structured neural network modeling” paradigm,which proposes aligning the model’s architecture with the intrinsic topology of financial products to enhance interpretability.Finally,this paper synthesizes the core challenges facing the field—including data quality,model robustness,and interpretability－and provides a forward-looking perspective on future research directions such as deep multi-modal fusion,causal inference,financial Large Language Models(LLMs),and explainable AI(XAI).

Key words: Financial forecasting, Time series, Portfolio management, Artificial intelligence, Machine learning, Deep learning

中图分类号:

TP39

陈夏伊. 人工智能在金融领域的数据预测方法综述[J]. 计算机科学, 2025, 52(12): 271-284. https://doi.org/10.11896/jsjkx.250700166

CHEN Xiayi. Survey of Data Prediction Methods Using Artificial Intelligence in the Financial Sector[J]. Computer Science, 2025, 52(12): 271-284. https://doi.org/10.11896/jsjkx.250700166

参考文献

[1]WANG G J,CHEN Y,ZHU Y,et al.Systemic risk predictionusing machine learning:Does network connectedness help prediction?[J].International Review of Financial Analysis,2024,93:103147.
[2]CHANG V,HAHM N,XU Q A,et al.Towards data and analytics driven B2B-banking for green finance:A cross-selling use case study[J].Technological Forecasting and Social Change,2024,206:123542.
[3]WANG S,CHI G.Cost-sensitive stacking ensemble learning for company financial distress prediction[J].Expert Systems with Applications,2024,255:124525.
[4]CHOU J S,CHEN K E.Optimizing investment portfolios with asequential ensemble of decision tree-based models and the FBI algorithm for efficient financial analysis[J].Applied Soft Computing,2024,158:111550.
[5]ÇELIK T B,Ι·CAN Ö,BULUT E.Extending machine learning prediction capabilities by explainable AI in financial time series prediction[J].Applied Soft Computing,2023,132:109876.
[6]KUO R J,CHIU T H.Hybrid of jellyfish and particle swarmoptimization algorithm-based support vector machine for stock market trend prediction[J].Applied Soft Computing,2024,154:111394.
[7]OLORUNNIMBE K,VIKTOR H.Deep learning in the stockmarket－a systematic survey of practice,backtesting,and applications[J].Artificial Intelligence Review,2023,56(3):2057-2109.
[8]MASINI R P,MEDEIROS M C,MENDES E F.Machine lear-ning advances for time series forecasting[J].Journal of economic surveys,2023,37(1):76-111.
[9]BALA R,SINGH R P.A dual-stage advanced deep learning algorithm for long-term and long-sequence prediction for multivariate financial time series[J].Applied Soft Computing,2022,126:109317.
[10]MUKHERJEE S,SADHUKHAN B,SARKAR N,et al.Stockmarket prediction using deep learning algorithms[J].CAAI Transactions on Intelligence Technology,2023,8(1):82-94.
[11]HUANG W C,CHEN C T,LEE C,et al.Attentive gated graph sequence neural network-based time-series information fusion for financial trading[J].Information Fusion,2023,91:261-276.
[12]LUO Q,BU J,XU W,et al.Stock market volatility prediction:Evidence from a new bagging model[J].International Review of Economics & Finance,2023,87:445-456.
[13]SONG Y,HUANG J,XU Y,et al.Multi-decomposition in deep learning models for futures price prediction[J].Expert Systems with Applications,2024,246:123171.
[14]WANG C,CHEN Y,ZHANG S,et al.Stock market index prediction using deep Transformer model[J].Expert Systems with Applications,2022,208:118128.
[15]CHENG D,YANG F,XIANG S,et al.Financial time seriesforecasting with multi-modality graph neural network[J].Pattern Recognition,2022,121:108218.
[16]JIN M,KOH H Y,WEN Q,et al.A survey on graph neural networks for time series:forecasting,classification,imputation,and anomaly detection[J].arXiv:2307.03759,2023.
[17]MA T,WANG W,CHEN Y.Attention is all you need:An interpretable transformer-based asset allocation approach[J].International Review of Financial Analysis,2023,90:102876.
[18]FISCHER T,STERLING M,LESSMANN S.Fx-spot predic-tions with state-of-the-art transformer and time embeddings[J].Expert Systems with Applications,2024,249:123538.
[19]LI B,CUI W,ZHANG L,et al.Difformer:Multi-resolutionaldifferencing transformer with dynamic ranging for time series analysis[J].IEEE transactions on pattern analysis and machine intelligence,2023,45(11):13586-13598.
[20]JANG J,SEONG N Y.Deep reinforcement learning for stock portfolio optimization by connecting with modern portfolio theory[J].Expert Systems with Applications,2023,218:119556.
[21]DU J.Mean-variance portfolio optimization with deep learning based-forecasts for cointegrated stocks[J].Expert Systems with Applications,2022,201:117005.
[22]DEZHKAM A,MANZURI M T.Forecasting stock market for an efficient portfolio by combining XGBoost and Hilbert-Huang transform[J].Engineering Applications of Artificial Intelligence,2023,118:105626.
[23]HENRIQUE B M,SOBREIRO V A,KIMURA H.Practical machine learning:Forecasting daily financial markets directions[J].Expert Systems with Applications,2023,233:120840.
[24]ZHOU Y,XIE C,WANG G J,et al.Forecasting cryptocurrency volatility:a novel framework based on the evolving multiscale graph neural network[J].Financial Innovation,2025,11(1):1-52.
[25]ASHTIANI M N,RAAHEMI B.News-based intelligent prediction of financial markets using text mining and machine lear-ning:A systematic literature review[J].Expert Systems with Applications,2023,217:119509.
[26]ALI A A,KHEDR A M,EL-BANNANY M,et al.A powerful predicting model for financial statement fraud based on optimized XGBoost ensemble learning technique[J].Applied Sciences,2023,13(4):2272.
[27]HTUN H H,BIEHL M,PETKOV N.Survey of feature selection and extraction techniques for stock market prediction[J].Financial Innovation,2023,9(1):26.
[28]ABDOLI M,AKBARI M,SHAHRABI J.Bagging supervisedautoencoder classifier for credit scoring[J].Expert Systems with Applications,2023,213:118991.
[29]SEBASTIÃO H,GODINHO P.Forecasting and trading cryptocurrencies with machine learning under changing market conditions[J].Financial Innovation,2021,7(1):3.
[30]CHI G,WANG S.Default risk prediction model of Chinese Listed Companies Based on xgboost [J].Journal of Systems Mana-gement,2024,33(3):735-754.
[31]YIN L,LI B,LI P,et al.Research on stock trend predictionmethod based on optimized random forest[J].CAAI Transactions on Intelligence Technology,2023,8(1):274-284.
[32]CHENG J,TIWARI S,KHALED D,et al.Forecasting Bitcoinprices using artificial intelligence:Combination of ML,SARIMA,and Facebook Prophet models[J].Technological Forecasting and Social Change,2024,198:122938.
[33]DAG A,DAG A Z,ASILKALKAN A,et al.A Tree Augmented Naïve Bayes-based methodology for classifying cryptocurrency trends[J].Journal of Business Research,2023,156:113522.
[34]SUN Z L,TANG J Y,FENG S,et al.Adaptive stock trading strategy based on deep reinforcement learning [J].Journal of Zhejiang University of technology,2024,52(2):188-195.
[35]PARK H J,KIM Y,KIM H Y.Stock market forecasting using a multi-task approach integrating long short-term memory and the random forest framework[J].Applied Soft Computing,2022,114:108106.
[36]EMMANUEL I,SUN Y,WANG Z.A machine learning-basedcredit risk prediction engine system using a stacked classifier and a filter-based feature selection method[J].Journal of Big Data,2024,11(1):23.
[37]ZHOU Y,SHEN L,BALLESTER L.A two-stage credit scoring model based on random forest:Evidence from Chinese small firms[J].International Review of Financial Analysis,2023,89:102755.
[38]DEL FAVA S,GUPTA R,PIERDZIOCH C,et al.Forecasting international financial stress:The role of climate risks[J].Journal of International Financial Markets,Institutions and Money,2024,92:101975.
[39]LYU X M,ZHANG R,et al.Personal credit risk assessment of online micro loan business－Based on DNN smoteenn extratrees combination model [J].Practice and Understanding of Mathematics,2023,53(7):14-21.
[40]BORUP D,CHRISTENSEN B J,MÜHLBACH N S,et al.Targeting predictors in random forest regression[J].International Journal of Forecasting,2023,39(2):841-868.
[41]WANG W,LIN W,WEN Y,et al.An interpretable intuitionistic fuzzy inference model for stock prediction[J].Expert Systems with Applications,2023,213:118908.
[42]MANTILLA P,DORMIDO-CANTO S.A novel feature engi-neering approach for high-frequency financial data[J].Engineering Applications of Artificial Intelligence,2023,125:106705.
[43]DING S,CUI T,BELLOTTI A G,et al.The role of feature importance in predicting corporate financial distress in pre and post COVID periods:Evidence from China[J].International Review of Financial Analysis,2023,90:102851.
[44]SAÂDAOUI F,RABBOUCH H.Financial forecasting improvement with LSTM-ARFIMA hybrid models and non-Gaussian distributions[J].Technological Forecasting and Social Change,2024,206:123539.
[45]IZADI M A,HAJIZADEH E.Time Series Prediction for Cryptocurrency Markets with Transformer and Parallel ConvolutionalNeural Networks[J].Applied Soft Computing,2025,113(3):113229.
[46]YιLDιRιM D C,TOROSLU I H,FIORE U.Forecasting directional movement of Forex data using LSTM with technical and macroeconomic indicators[J].Financial innovation,2021,7(1):1.
[47]ZHAO Y,ZHANG W,LIU X.Grid search with a weighted error function:Hyper-parameter optimization for financial time series forecasting[J].Applied Soft Computing,2024,154:111362.
[48]GAJAMANNAGE K,PARK Y,JAYATHILAKE D I.Real-time forecasting of time series in financial markets using sequentially trained dual-LSTMs[J].Expert Systems with Applications,2023,223:119879.
[49]DONG Z,ZHOU Y.A novel hybrid model for financial forecasting based on CEEMDAN-SE and ARIMA-CNN-LSTM[J].Mathematics,2024,12(16):2434.
[50]ZHAO Y,YANG G.Deep Learning-based Integrated Frame-work for stock price movement prediction[J].Applied Soft Computing,2023,133:109921.
[51]HUANG Y,WANG Z,JIANG C.Diagnosis with incompletemulti-view data:A variational deep financial distress prediction method[J].Technological Forecasting and Social Change,2024,201:123269.
[52]ZHANG P,HARRIS R D F,ZHENG J.GNN-based socialmedia sentiment analysis for stock market forecasting and trading[J].Expert Systems with Applications,2025:128425.
[53]LAZCANO A,HERRERA P J,MONGE M.A combined model based on recurrent neural networks and graph convolutional networks for financial time series forecasting[J].Mathematics,2023,11(1):224.
[54]HUANG Z,LI K,JIANG Y,et al.Graph Relearn Network:Re-ducing performance variance and improving prediction accuracy of graph neural networks[J].Knowledge-Based Systems,2024,301:112311.
[55]LEE S,CHO P.Graph-Based Stock Volatility Forecasting with Effective Transfer Entropy and Hurst-Based Regime Adaptation[J].Fractal and Fractional,2025,9(6):339.
[56]TAO Z,WU W,WANG J.Series decomposition Transformerwith period-correlation for stock market index prediction[J].Expert Systems with Applications,2024,237:121424.
[57]YANG S.Deep reinforcement learning for portfolio management[J].Knowledge-Based Systems,2023,278:110905.
[58]HUANG Y,ZHOU C,CUI K,et al.A multi-agent reinforcement learning framework for optimizing financial trading strategies based on TimesNet[J].Expert Systems with Applications,2024,237:121502.
[59]LI S Y,ZHONG X Y,LI K Y,et al.Research on strategy teaching based on multi-layer graph relationship and reinforcement learning [j].Computer Engineering,2025,51(3):122-130.
[60]MISHRA A K,RENGANATHAN J,GUPTA A.Volatilityforecasting and assessing risk of financial markets using multi-transformer neural network based architecture[J].Engineering Applications of Artificial Intelligence,2024,133:108223.
[61]HAQ A U,ZEB A,LEI Z,et al.Forecasting daily stock trend using multi-filter feature selection and deep learning[J].Expert Systems with Applications,2021,168:114444.
[62]KONG L,ZHENG G,BRINTRUP A.A federated machinelearning approach for order-level risk prediction in supply chain financing[J].International Journal of Production Economics,2024,268:109095.
[63]AYITEY J M,APPIAHENE P,APPIAH O,et al.Forex market forecasting using machine learning:Systematic Literature Review and meta-analysis[J].Journal of Big Data,2023,10(1):9.
[64]ZHANG Y,WANG G,ZHOU T,et al.Takagi-Sugeno-Kangfuzzy system fusion:A survey at hierarchical,wide and stacked levels[J].Information fusion,2024,101:101977.
[65]YANG M,LIM M K,QU Y,et al.Deep neural networks with L1 and L2 regularization for high dimensional corporate credit risk prediction[J].Expert Systems with Applications,2023,213:118873.
[66]OYEDELE A A,AJAYI A O,OYEDELE L O,et al.Perfor-mance evaluation of deep learning and boosted trees for cryptocurrency closing price prediction[J].Expert Systems with Applications,2023,213:119233.
[67]LONG W,GAO J,BAI K,et al.A hybrid model for stock price prediction based on multi-view heterogeneous data[J].Financial Innovation,2024,10(1):48.
[68]SONG Y,CAI C,MA D,et al.Modelling and forecasting high-frequency data with jumps based on a hybrid nonparametric regression and LSTM model[J].Expert Systems with Applications,2024,237:121527.
[69]ZHANG Q,QIN C,ZHANG Y,et al.Transformer-based attention network for stock movement prediction[J].Expert Systems with Applications,2022,202:117239.
[70]EDALATPANAH S A,HASSANI F S,SMARANDACHE F,et al.A hybrid time series forecasting method based on neutrosophic logic with applications in financial issues[J].Engineering Applications of Artificial Intelligence,2024,129:107531.
[71]ROOSTAEE M R,ABIN A A.Forecasting financial signal for automated trading:An interpretable approach[J].Expert Systems with Applications,2023,211:118570.
[72]RAZA S A,KHAN K A,BENKRAIEM R,et al.The impor-tance of climate policy uncertainty in forecasting the green,clean and sustainable financial markets volatility[J].International Review of Financial Analysis,2024,91:102984.
[73]YU B,LI C,MIRZA N,et al.Forecasting credit ratings of decarbonized firms:Comparative assessment of machine learning models[J].Technological Forecasting and Social Change,2022,174:121255.
[74]SUN Z Q,LIU X L.Research on risk taking,trend of net inte-rest margin and stable development of banks－experimental evidence based on panel data of commercial banks from 2009 to 2019 [J].Financial Theory and Practice,2023,44(3):18-26.
[75]CHAUDHARI K,THAKKAR A.Neural network systems with an integrated coefficient of variation-based feature selection for stock price and trend prediction[J].Expert Systems with Applications,2023,219:119527.
[76]LU M,XU X.TRNN:An efficient time-series recurrent neural network for stock price prediction[J].Information Sciences,2024,657:119951.
[77]HU M,TAN Z,LIU B,et al.Graph Portfolio:High-Frequency Factor Predictors via Heterogeneous Continual GNNs[J].IEEE Transactions on Knowledge and Data Engineering,2025,37(7):4104-4116.
[78]WANG Y,ANDREEVA G,Martin-Barragan B.Machine lear-ning approaches to forecasting cryptocurrency volatility:Consi-dering internal and external determinants[J].International Review of Financial Analysis,2023,90:102914.
[79]MD A Q,KAPOOR S,AV C J,et al.Novel optimization ap-proach for stock price forecasting using multi-layered sequential LSTM[J].Applied Soft Computing,2023,134:109830.
[80]SHAO Z,YAO X,CHEN F,et al.Revisiting time-varying dynamics in stock market forecasting:A multi-source sentiment analysis approach with large language model[J].Decision Support Systems,2025,190:114362.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed