Scholarlink

The future is unpredictable and unknowable, but possible ways exist to make predictions and get benefits securely. One such possibility is using AI and DL to forecast the stock market. The equity market’s dynamic nature and intrinsic volatility make this area more interesting for researchers. The fluctuations in the stock market depend on many factors that make future predictions more difficult using simple traditional statistical models. Therefore, this research suggests two optimized DL approaches for the prediction of the closing price of the stock one week prior. In the current study, RNN and LSTM with deep hyperparameter tuning have been applied to the recent dataset of three IT companies to predict their future closing price. The performance of these approaches has been evaluated using four evaluation metrics including MAPE, R2, MAE, and RMSE. As per the results achieved in this study, it has been interpreted that the LSTM shows good and acceptable results compared to RNN and it appears to be the suitable choice for making future investment decisions.

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

Future Price Prediction of IT Sector Companies Using Optimized Deep Learning Approaches

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Umar Bashir,
  • Kuljeet Singh,
  • Megha Raina,
  • Vibhakar Mansotra

摘要

The future is unpredictable and unknowable, but possible ways exist to make predictions and get benefits securely. One such possibility is using AI and DL to forecast the stock market. The equity market’s dynamic nature and intrinsic volatility make this area more interesting for researchers. The fluctuations in the stock market depend on many factors that make future predictions more difficult using simple traditional statistical models. Therefore, this research suggests two optimized DL approaches for the prediction of the closing price of the stock one week prior. In the current study, RNN and LSTM with deep hyperparameter tuning have been applied to the recent dataset of three IT companies to predict their future closing price. The performance of these approaches has been evaluated using four evaluation metrics including MAPE, R2, MAE, and RMSE. As per the results achieved in this study, it has been interpreted that the LSTM shows good and acceptable results compared to RNN and it appears to be the suitable choice for making future investment decisions.