Progresses of Machine Learning in Stock Market Prediction: A Comprehensive Model Comparison

Zheng Qin

2024

Abstract

Stock price prediction is a crucial technique in investment, but traditional approaches can hardly get high accuracy and precision on predicting as they cannot considered about the various factors in the complex market. As a result, more advanced method is desired by the market. In this review, several major methods of machine learning of artificial intelligence are referred, such as Linear Regression, Random Forest, Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM). And the basic workflow behind the application of these methods is also introduced. Based on the cases of different models in predicting stock prices, the advantages and disadvantages of various machine learning models in forecasting are illustrated and compared. Traditional models like Linear Regression offer simplicity but are limited by their inability to capture non-linear relationships. In contrast, Random Forest models improve prediction accuracy through ensemble methods but face challenges like overfitting and lack of temporal modelling capabilities. Advanced models such as RNNs and LSTMs excel in capturing complex temporal dependencies, making them more effective for stock price forecasting. This review can provide a good reference for scholars in the intersection of finance and AI.

Paper Citation

in Harvard Style

Qin Z. (2024). Progresses of Machine Learning in Stock Market Prediction: A Comprehensive Model Comparison. In Proceedings of the 2nd International Conference on Data Analysis and Machine Learning - Volume 1: DAML; ISBN 978-989-758-754-2, SciTePress, pages 96-100. DOI: 10.5220/0013509100004619

in Bibtex Style

@conference{daml24,
author={Zheng Qin},
title={Progresses of Machine Learning in Stock Market Prediction: A Comprehensive Model Comparison},
booktitle={Proceedings of the 2nd International Conference on Data Analysis and Machine Learning - Volume 1: DAML},
year={2024},
pages={96-100},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013509100004619},
isbn={978-989-758-754-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 2nd International Conference on Data Analysis and Machine Learning - Volume 1: DAML
TI - Progresses of Machine Learning in Stock Market Prediction: A Comprehensive Model Comparison
SN - 978-989-758-754-2
AU - Qin Z.
PY - 2024
SP - 96
EP - 100
DO - 10.5220/0013509100004619
PB - SciTePress