Scholarlink

This article presents autoregressive integrated moving average (ARIMA) and long short-term memory (LSTM) network-based machine learning models for precise prediction of Dow Jones U.S. Auto Manufacturers (DJUSAM) Index using python-spyder platform. The stock closing data of past 3 months was used for training of the proposed models. The accuracy of models was analysed by predicting next 7 days of stock closing data. The magnitude of errors was measured in terms of mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE). The results highlight that the magnitude of errors increases with duration of prediction. The results show that minimal MAPE values in order of 2.77% and 2.92% were obtained using ARIMA and LSTM models, respectively. The ARIMA model provided comparatively better results compared to LSTM model giving lower error responses.

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

Analysing Various Machine Learning Models for Auto-Index Prediction

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Neeraj Sengar,
  • Amir Shaikh,
  • Syed Farrukh Rasheed,
  • Gagan Bansal,
  • Sankula Madhava,
  • Kanuri Pranay,
  • Sai Jayanth Bobbili,
  • Shaik Tajuddin,
  • Patibandla Tejaswantt

摘要

This article presents autoregressive integrated moving average (ARIMA) and long short-term memory (LSTM) network-based machine learning models for precise prediction of Dow Jones U.S. Auto Manufacturers (DJUSAM) Index using python-spyder platform. The stock closing data of past 3 months was used for training of the proposed models. The accuracy of models was analysed by predicting next 7 days of stock closing data. The magnitude of errors was measured in terms of mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE). The results highlight that the magnitude of errors increases with duration of prediction. The results show that minimal MAPE values in order of 2.77% and 2.92% were obtained using ARIMA and LSTM models, respectively. The ARIMA model provided comparatively better results compared to LSTM model giving lower error responses.