Scholarlink

The prediction of stock market indices occupies a vital place with investors, portfolio managers, and stock market regulators. It is a difficult problem due to multi-noise, non-linearity, and stock price stochastic. These features of stocks prevent most conventional forecasting models from extracting useful information from historical data. In this study, we suggest a hybrid model that mixes Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNN) with a self-attention mechanism to predict future stock market index prices. Our approach is based on historical closing prices from the standard and poor’s 500 (SP&500). The predictive performance of our model is assessed by Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), and Mean Absolute Error (MAE).

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

Predicting the Index of the Stock Market Employing an LSTM-CNN Hybrid Model with an Attention Mechanism

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Zakaria Al Bakkari,
  • Ikram El Azami,
  • Adil El Makrani,
  • Moulay Hafid Aabidi

摘要

The prediction of stock market indices occupies a vital place with investors, portfolio managers, and stock market regulators. It is a difficult problem due to multi-noise, non-linearity, and stock price stochastic. These features of stocks prevent most conventional forecasting models from extracting useful information from historical data. In this study, we suggest a hybrid model that mixes Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNN) with a self-attention mechanism to predict future stock market index prices. Our approach is based on historical closing prices from the standard and poor’s 500 (SP&500). The predictive performance of our model is assessed by Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), and Mean Absolute Error (MAE).