Multi-step-ahead copper price forecasting using a two-phase architecture based on an improved LSTM with novel input strategy and error correction DOI

Hongyuan Luo,

Deyun Wang,

Jinhua Cheng

et al.

Resources Policy, Journal Year: 2022, Volume and Issue: 79, P. 102962 - 102962

Published: Sept. 10, 2022

Language: Английский

FE-RNN: A fuzzy embedded recurrent neural network for improving interpretability of underlying neural network DOI Creative Commons

James Chee Min Tan,

Qi Cao, Chai Quek

et al.

Information Sciences, Journal Year: 2024, Volume and Issue: 663, P. 120276 - 120276

Published: Feb. 9, 2024

Deep learning enables effective predictions. But deep structures face some challenges on human interpretability compared to conventional techniques, e.g., fuzzy inference systems. It motivates more research works alleviate the black box nature of with performance maintained. This paper proposes a fuzzy-embedded recurrent neural network (FE-RNN) improve underlying networks. is parallel structure comprising an RNN and Pseudo Outer-Product based Fuzzy Neural Network (POPFNN) that share common set input output linguistic concepts. The processes undertaken are associated by using rules in embedded POPFNN. IF-THEN provide better process hybrid allows realisation data driven implication modelling entailment within networks (FNN) structure. FE-RNN obtains consistent results than other FNN experiment Mackey-Glass dataset. achieves about 99% correlation for forecasting prices market indexes. Its also discussed. then acts as prediction tool financial trading system forecast-assisted technical indicators optimised Genetic Algorithms. outperforms benchmark strategies experiments.

Language: Английский

Citations

11
Towards Economic Sustainability: A Comprehensive Review of Artificial Intelligence and Machine Learning Techniques in Improving the Accuracy of Stock Market Movements DOI Creative Commons

Atoosa Rezaei,

Iheb Abdellatif,

Amjad Umar

et al.

International Journal of Financial Studies, Journal Year: 2025, Volume and Issue: 13(1), P. 28 - 28

Published: Feb. 25, 2025

Accurately predicting stock market movements remains a critical challenge in finance, driven by the increasing role of algorithmic trading and centrality financial markets economic sustainability. This study examines incorporation artificial intelligence (AI) machine learning (ML) technologies to address gaps identifying predictive factors, integrating diverse data sources, optimizing methodologies. Employing systematic review, recent advancements ML techniques like deep learning, ensemble methods, neural networks are analyzed, alongside emerging sources such as traders’ sentiment real-time indicators. Results highlight potential unified datasets adaptive models enhance prediction accuracy while overcoming volatility heterogeneity. The research underscores necessity innovative advanced develop robust adaptable forecasting frameworks. These findings offer valuable insights for academics professionals, paving way more reliable that can decision-making dynamic environments. contributes advancing sustainability proposing methodologies align with complexities rapid evolution modern markets.

Language: Английский

Citations

1
McVCsB: A new hybrid deep learning network for stock index prediction DOI
Chenhao Cui, Peiwan Wang, Yong Li

et al.

Expert Systems with Applications, Journal Year: 2023, Volume and Issue: 232, P. 120902 - 120902

Published: June 26, 2023

Language: Английский

Citations

22
Effective LSTMs with seasonal-trend decomposition and adaptive learning and niching-based backtracking search algorithm for time series forecasting DOI
Yuhan Wu, Xiyu Meng, Junru Zhang

et al.

Expert Systems with Applications, Journal Year: 2023, Volume and Issue: 236, P. 121202 - 121202

Published: Aug. 26, 2023

Language: Английский

Citations

22
A novel hybrid model for stock price forecasting integrating Encoder Forest and Informer DOI
Shangsheng Ren, Xu Wang,

Xu Zhou

et al.

Expert Systems with Applications, Journal Year: 2023, Volume and Issue: 234, P. 121080 - 121080

Published: Aug. 4, 2023

Language: Английский

Citations

21
An interval constraint-based trading strategy with social sentiment for the stock market DOI Creative Commons
Mingchen Li, Kun Yang, Wencan Lin

et al.

Financial Innovation, Journal Year: 2024, Volume and Issue: 10(1)

Published: Feb. 10, 2024

Abstract Developing effective strategies to earn excess returns in the stock market is a cutting-edge topic field of economics. At same time, price forecasting that supports trading considered one most challenging tasks. Therefore, this study analyzes and extracts news media data, expert comments, social opinion pandemic text data using natural language processing, then combines with deep learning model forecast future patterns based on historical prices. An interval constraint-based strategy constructed. Using from several typical stocks Chinese during COVID-19 period, empirical studies simulations show, first, sentiment composite index can improve accuracy forecasting. Second, proposed approach effectively enhance thus, assist investors decision-making.

Language: Английский

Citations

8
Can Denoising Enhance Prediction Accuracy of Learning Models? A Case of Wavelet Decomposition Approach DOI Creative Commons

C. Tamilselvi,

Md Yeasin, Ranjit Kumar Paul

et al.

Forecasting, Journal Year: 2024, Volume and Issue: 6(1), P. 81 - 99

Published: Jan. 16, 2024

Denoising is an integral part of the data pre-processing pipeline that often works in conjunction with model development for enhancing quality data, improving accuracy, preventing overfitting, and contributing to overall robustness predictive models. Algorithms based on a combination wavelet deep learning, machine stochastic have been proposed. The denoised series are fitted various benchmark models, including long short-term memory (LSTM), support vector regression (SVR), artificial neural network (ANN), autoregressive integrated moving average (ARIMA) effectiveness wavelet-based denoising approach was investigated monthly wholesale price three major spices (turmeric, coriander, cumin) markets India. performance these models assessed using root mean square error (RMSE), absolute percentage (MAPE), (MAE). LSTM Haar filter at level 6 emerged as robust choice accurate predictions across all spices. It found had significant gain accuracy than by more 30% metrics. results clearly highlighted efficacy forecasting.

Language: Английский

Citations

6
TimeSQL: Improving multivariate time series forecasting with multi-scale patching and smooth quadratic loss DOI

Site Mo,

Haoxin Wang,

Bixiong Li

et al.

Information Sciences, Journal Year: 2024, Volume and Issue: 671, P. 120652 - 120652

Published: April 25, 2024

Language: Английский

Citations

6
Multi-step Carbon Emissions Forecasting Model for Industrial Process Based on a New Strategy and Machine Learning Methods DOI

Yusha Hu,

Yi Man, Jingzheng Ren

et al.

Process Safety and Environmental Protection, Journal Year: 2024, Volume and Issue: 187, P. 1213 - 1233

Published: May 14, 2024

Language: Английский

Citations

6
Integrating Navier-Stokes equation and neoteric iForest-BorutaShap-Facebook’s prophet framework for stock market prediction: An application in Indian context DOI
Indranil Ghosh, Tamal Datta Chaudhuri

Expert Systems with Applications, Journal Year: 2022, Volume and Issue: 210, P. 118391 - 118391

Published: Aug. 4, 2022

Language: Английский

Citations

23