Artificial intelligence techniques in financial trading: A systematic literature review

Journal of King Saud University - Computer and Information Sciences, Journal Year: 2024, Volume and Issue: 36(3), P. 102015 - 102015

Published: March 1, 2024

Artificial Intelligence (AI) approaches have been increasingly used in financial markets as technology advances. In this research paper, we conduct a Systematic Literature Review (SLR) that studies trading through AI techniques. It reviews 143 articles implemented techniques markets. Accordingly, it presents several findings and observations after reviewing the papers from following perspectives: market asset type, analysis type considered along with technique, utilized market, estimation performance metrics of proposed models. The selected were published between 2015 2023, review addresses four RQs. After analyzing articles, observed 8 building predictive Moreover, found technical is more adopted compared to fundamental analysis. Furthermore, 16% entirely automate process. addition, identified 40 different are standalone hybrid Among these techniques, deep learning most frequently Building prediction models for using promising field research, academics already deployed machine As result evaluation, provide recommendations guidance researchers.

Language: Английский

---

A Comprehensive Review of Deep Learning: Architectures, Recent Advances, and Applications

Information, Journal Year: 2024, Volume and Issue: 15(12), P. 755 - 755

Published: Nov. 27, 2024

Deep learning (DL) has become a core component of modern artificial intelligence (AI), driving significant advancements across diverse fields by facilitating the analysis complex systems, from protein folding in biology to molecular discovery chemistry and particle interactions physics. However, field deep is constantly evolving, with recent innovations both architectures applications. Therefore, this paper provides comprehensive review DL advances, covering evolution applications foundational models like convolutional neural networks (CNNs) Recurrent Neural Networks (RNNs), as well such transformers, generative adversarial (GANs), capsule networks, graph (GNNs). Additionally, discusses novel training techniques, including self-supervised learning, federated reinforcement which further enhance capabilities models. By synthesizing developments identifying current challenges, insights into state art future directions research, offering valuable guidance for researchers industry experts.

Language: Английский

---

Stock market prediction with time series data and news headlines: a stacking ensemble approach

Journal of Intelligent Information Systems, Journal Year: 2023, Volume and Issue: 62(1), P. 27 - 56

Published: July 23, 2023

Language: Английский

---

Random kernel k-nearest neighbors regression

Frontiers in Big Data, Journal Year: 2024, Volume and Issue: 7

Published: July 1, 2024

The k-nearest neighbors (KNN) regression method, known for its nonparametric nature, is highly valued simplicity and effectiveness in handling complex structured data, particularly big data contexts. However, this method susceptible to overfitting fit discontinuity, which present significant challenges. This paper introduces the random kernel (RK-KNN) as a novel approach that well-suited applications. It integrates smoothing with bootstrap sampling enhance prediction accuracy robustness of model. aggregates multiple predictions using from training dataset selects subsets input variables KNN (K-KNN). A comprehensive evaluation RK-KNN on 15 diverse datasets, employing various functions including Gaussian Epanechnikov, demonstrates superior performance. When compared standard (R-KNN) models, it significantly reduces root mean square error (RMSE) absolute error, well improving R-squared values. variant employs specific function yielding lowest RMSE will be benchmarked against state-of-the-art methods, support vector regression, artificial neural networks, forests.

Language: Английский

---

Encoder–Decoder Based LSTM and GRU Architectures for Stocks and Cryptocurrency Prediction

Journal of risk and financial management, Journal Year: 2024, Volume and Issue: 17(5), P. 200 - 200

Published: May 12, 2024

This work addresses the intricate task of predicting prices diverse financial assets, including stocks, indices, and cryptocurrencies, each exhibiting distinct characteristics behaviors under varied market conditions. To tackle challenge effectively, novel encoder–decoder architectures, AE-LSTM AE-GRU, integrating principle with LSTM GRU, are designed. The experimentation involves multiple activation functions hyperparameter tuning. With extensive enhancements applied to AE-LSTM, proposed AE-GRU architecture still demonstrates significant superiority in forecasting annual volatile assets from sectors mentioned above. Thus, emerges as a superior choice for price prediction across fluctuating scenarios by extracting important non-linear features data retaining long-term context past observations.

Language: Английский

---

New deep recurrent hybrid artificial neural network for forecasting seasonal time series

Granular Computing, Journal Year: 2024, Volume and Issue: 9(1)

Published: Jan. 10, 2024

Language: Английский

---

Deep learning–based short-term water demand forecasting in urban areas: A hybrid multichannel model

AQUA - Water Infrastructure Ecosystems and Society, Journal Year: 2024, Volume and Issue: 73(3), P. 380 - 395

Published: Feb. 28, 2024

Abstract Forecasting short-term water demands is one of the most critical needs operating companies urban distribution networks. Water have a time series nature, and various factors affect their variations patterns, which make it difficult to forecast. In this study, we first implemented hybrid model convolutional neural networks (CNNs) recurrent (RNNs) forecast demand. These models include combination CNN with simple RNN (CNN-Simple RNN), gate unit (CNN-GRU), long memory (CNN-LSTM). Then, increased number channels achieve higher accuracy. The accuracy up four. evaluation metrics show that CNN-GRU superior other models. Ultimately, four-channel demonstrated highest accuracy, achieving mean absolute percentage error (MAPE) 1.65% for 24-h forecasting horizon. effects horizon on results were also investigated. MAPE 1-h 1.06% in CNN-GRU, its value decreases amount

Language: Английский

---

Portfolio Optimization with Long-Short Term Memory Deep Learning (LSTM)

Deleted Journal, Journal Year: 2025, Volume and Issue: 20(2), P. 1 - 14

Published: Feb. 27, 2025

The objective is a methodology for weighting financial assets in an investment portfolio. It contrasted by the components of Dow Jones Industrial Average (DJIA). For this purpose, portfolios with horizons between 1 and 2 years are studied using Long-Short Term Memory (LSTM) optimization. best portfolio was horizon 1.5 years. neural network trained 1,000 observations more than 2,777 simulated. model outperforms DJIA 73% to 85%, geometric mean annual return differential 3.7% 5%. history used allocate 2008 2021. recommended that be conjunction another selecting assets. conclusions limited make up DJIA. Mostly literature, networks short term; paper contrasts long term, seeking weigh not future asset prices. conclusion LSTM can

Language: Английский

---

Leveraging deep learning for risk prediction and resilience in supply chains: insights from critical industries

Journal Of Big Data, Journal Year: 2025, Volume and Issue: 12(1)

Published: April 17, 2025

Language: Английский

---

An Accurate Multiple Data Based Stock Prediction and Sentiment Analysis Using Synergic Deep Info Convolutional Neural Network

Computational Economics, Journal Year: 2025, Volume and Issue: unknown

Published: April 21, 2025

Language: Английский

---