Scholarlink

<p>Foreign exchange trend prediction is a popular research topic since traders and researchers have sought to outperform the market and generate steady profits. Studies have been conducted to predict the direction of trends using decision trees. To improve the performance of decision trees in forex prediction, this study proposes an ensemble approach that integrates convolutional neural network (CNN), bagging and boosting techniques. The Great British Pound (GBP) against the Japanese yen (JPY) currency pair is used for testing, which pits the British pound against the Japanese yen (GBP/JPY). CNN is used for feature extraction. The ensemble combines several decision trees to produce better predictive performance. The feature set used includes technical indicators and the open, low, high, and close of one-hour prices of the GBP against the JPY. To demonstrate the effectiveness of our proposed model, we compared the prediction performance with previously published algorithms such as random forest, K- nearest neighbors, and extreme gradient boosting. We obtain an accuracy of 0.923, AUC of 0.944, Precision of 0.845, and Recall of 0.959 after cross validation. The results of the study show that the proposed method outperforms other methods.</p>

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

Ensemble Learning for Foreign Exchange Market Trend Prediction

  • 期刊论文

  • 发表日期：

    2025年9月22日

查看全文
  • Ekla Njoki,
  • Jael Sanyanda Wekesa,
  • Denis Gitari Njagi

摘要

Foreign exchange trend prediction is a popular research topic since traders and researchers have sought to outperform the market and generate steady profits. Studies have been conducted to predict the direction of trends using decision trees. To improve the performance of decision trees in forex prediction, this study proposes an ensemble approach that integrates convolutional neural network (CNN), bagging and boosting techniques. The Great British Pound (GBP) against the Japanese yen (JPY) currency pair is used for testing, which pits the British pound against the Japanese yen (GBP/JPY). CNN is used for feature extraction. The ensemble combines several decision trees to produce better predictive performance. The feature set used includes technical indicators and the open, low, high, and close of one-hour prices of the GBP against the JPY. To demonstrate the effectiveness of our proposed model, we compared the prediction performance with previously published algorithms such as random forest, K- nearest neighbors, and extreme gradient boosting. We obtain an accuracy of 0.923, AUC of 0.944, Precision of 0.845, and Recall of 0.959 after cross validation. The results of the study show that the proposed method outperforms other methods.