Scholarlink

The importance of stock price prediction lies in its potential to inform investment decisions on risk management. It reflects the collective perceptions and expectations of investors regarding the company’s performance and future prospects. This paper introduces an adaptive random forest-LSTM ensemble for stock price prediction, combining the strengths of random forest and long short-term memory (LSTM) models. This paper explores the application of machine learning (ML) and deep learning (DL) techniques for stock price prediction. Initially, four ML algorithms multiple linear regression (MLR), support vector regression (SVR), decision tree regression (DTR), and random forest regression (RFR) were employed, with RFR exhibiting a notable R-squared value of 0.98. Subsequently, three DL techniques, namely, RNN, LSTM, and GRU, were applied, with LSTM yielding favorable results. Later, a novel adaptive ensemble model, amalgamating the strengths of ML and DL methods, showcases robust performance with an R-squared value of 0.99 and a lower mean squared error (MSE) of 121. The proposed adaptive ensemble model is a promising approach for making more precise stock price predictions, potentially improving decision-making in dynamic market situations. It outperformed the existing conventional methods for stock price prediction.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Adaptive Ensemble of ML Regressors and LSTM for Stock Price Prediction

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • K. Suresh Babu,
  • Keshetti Sreekala,
  • S. Sushma,
  • T. V. Sai Krishna,
  • A. Lakshmanarao

摘要

The importance of stock price prediction lies in its potential to inform investment decisions on risk management. It reflects the collective perceptions and expectations of investors regarding the company’s performance and future prospects. This paper introduces an adaptive random forest-LSTM ensemble for stock price prediction, combining the strengths of random forest and long short-term memory (LSTM) models. This paper explores the application of machine learning (ML) and deep learning (DL) techniques for stock price prediction. Initially, four ML algorithms multiple linear regression (MLR), support vector regression (SVR), decision tree regression (DTR), and random forest regression (RFR) were employed, with RFR exhibiting a notable R-squared value of 0.98. Subsequently, three DL techniques, namely, RNN, LSTM, and GRU, were applied, with LSTM yielding favorable results. Later, a novel adaptive ensemble model, amalgamating the strengths of ML and DL methods, showcases robust performance with an R-squared value of 0.99 and a lower mean squared error (MSE) of 121. The proposed adaptive ensemble model is a promising approach for making more precise stock price predictions, potentially improving decision-making in dynamic market situations. It outperformed the existing conventional methods for stock price prediction.