Scholarlink

<p>Machine learning methods with technical analysis indicators have gained traction, aiming to enhance decision-making within energy trading companies. This research assesses the effectiveness of several machine learning models, including linear regression, logistic regression, XGBoost, light gradient boosting machine, random forests, Gaussian naive Bayes and support vector machines considering Sharpe index, accuracy, F1 score, Mean Squared Error, and Root Mean Squared Error metrics. Due to this assessment, we propose a hybrid model that shows a 70% accuracy in generating buy and sell signals, achieving a 13 logarithmic return over 350 trading days. We used Brazilian and Italian electricity market datasets to check the integrity of the proposed model. This research contributes to more informed and effective decision-making in the power trading market.</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

Technical Analysis and Machine Learning Applied to the Short-Term Electricity Trading Market: Italian and Brazilian Cases

  • 期刊论文

  • 发表日期：

    2025年4月25日

查看全文
  • Raphael Paulo Beal Piovezan,
  • Pedro Paulo de Andrade Junior,
  • Sérgio Luciano Ávila,
  • Erinaldo Farias dos Santos

摘要

Machine learning methods with technical analysis indicators have gained traction, aiming to enhance decision-making within energy trading companies. This research assesses the effectiveness of several machine learning models, including linear regression, logistic regression, XGBoost, light gradient boosting machine, random forests, Gaussian naive Bayes and support vector machines considering Sharpe index, accuracy, F1 score, Mean Squared Error, and Root Mean Squared Error metrics. Due to this assessment, we propose a hybrid model that shows a 70% accuracy in generating buy and sell signals, achieving a 13 logarithmic return over 350 trading days. We used Brazilian and Italian electricity market datasets to check the integrity of the proposed model. This research contributes to more informed and effective decision-making in the power trading market.